Computer Vision and Pattern Recognition (cs.CV)

Abstract: We propose a novel-view augmentation (NOVA) strategy to train NeRFs for photo-realistic 3D composition of dynamic objects in a static scene. Compared to prior work, our framework significantly reduces blending artifacts when inserting multiple dynamic objects into a 3D scene at novel views and times; achieves comparable PSNR without the need for additional ground truth modalities like optical flow; and overall provides ease, flexibility, and scalability in neural composition. Our codebase is on GitHub.

Abstract: High-Definition (HD) maps are essential for the safety of autonomous driving systems. While existing techniques employ camera images and onboard sensors to generate vectorized high-precision maps, they are constrained by their reliance on single-frame input. This approach limits their stability and performance in complex scenarios such as occlusions, largely due to the absence of temporal information. Moreover, their performance diminishes when applied to broader perception ranges. In this paper, we present StreamMapNet, a novel online mapping pipeline adept at long-sequence temporal modeling of videos. StreamMapNet employs multi-point attention and temporal information which empowers the construction of large-range local HD maps with high stability and further addresses the limitations of existing methods. Furthermore, we critically examine widely used online HD Map construction benchmark and datasets, Argoverse2 and nuScenes, revealing significant bias in the existing evaluation protocols. We propose to resplit the benchmarks according to geographical spans, promoting fair and precise evaluations. Experimental results validate that StreamMapNet significantly outperforms existing methods across all settings while maintaining an online inference speed of $14.2$ FPS.

Abstract: Existing K-nearest neighbor (KNN) retrieval-based methods usually conduct industrial anomaly detection in two stages: obtain feature representations with a pre-trained CNN model and perform distance measures for defect detection. However, the features are not fully exploited as they ignore domain bias and the difference of local density in feature space, which limits the detection performance. In this paper, we propose Reducing Biases (REB) in representation by considering the domain bias of the pre-trained model and building a self-supervised learning task for better domain adaption with a defect generation strategy (DefectMaker) imitating the natural defects. Additionally, we propose a local density KNN (LDKNN) to reduce the local density bias and obtain effective anomaly detection. We achieve a promising result of 99.5\% AUROC on the widely used MVTec AD benchmark. We also achieve 88.0\% AUROC on the challenging MVTec LOCO AD dataset and bring an improvement of 4.7\% AUROC to the state-of-the-art result. All results are obtained with smaller backbone networks such as Vgg11 and Resnet18, which indicates the effectiveness and efficiency of REB for practical industrial applications.

Abstract: Handling entirely unknown data is a challenge for any deployed classifier. Classification models are typically trained on a static pre-defined dataset and are kept in the dark for the open unassigned feature space. As a result, they struggle to deal with out-of-distribution data during inference. Addressing this task on the class-level is termed open-set recognition (OSR). However, most OSR methods are inherently limited, as they train closed-set classifiers and only adapt the downstream predictions to OSR. This work presents LORD, a framework to Leverage Open-set Recognition by exploiting unknown Data. LORD explicitly models open space during classifier training and provides a systematic evaluation for such approaches. We identify three model-agnostic training strategies that exploit background data and applied them to well-established classifiers. Due to LORD's extensive evaluation protocol, we consistently demonstrate improved recognition of unknown data. The benchmarks facilitate in-depth analysis across various requirement levels. To mitigate dependency on extensive and costly background datasets, we explore mixup as an off-the-shelf data generation technique. Our experiments highlight mixup's effectiveness as a substitute for background datasets. Lightweight constraints on mixup synthesis further improve OSR performance.

Abstract: Cross-modal alignment is one key challenge for Vision-and-Language Navigation (VLN). Most existing studies concentrate on mapping the global instruction or single sub-instruction to the corresponding trajectory. However, another critical problem of achieving fine-grained alignment at the entity level is seldom considered. To address this problem, we propose a novel Grounded Entity-Landmark Adaptive (GELA) pre-training paradigm for VLN tasks. To achieve the adaptive pre-training paradigm, we first introduce grounded entity-landmark human annotations into the Room-to-Room (R2R) dataset, named GEL-R2R. Additionally, we adopt three grounded entity-landmark adaptive pre-training objectives: 1) entity phrase prediction, 2) landmark bounding box prediction, and 3) entity-landmark semantic alignment, which explicitly supervise the learning of fine-grained cross-modal alignment between entity phrases and environment landmarks. Finally, we validate our model on two downstream benchmarks: VLN with descriptive instructions (R2R) and dialogue instructions (CVDN). The comprehensive experiments show that our GELA model achieves state-of-the-art results on both tasks, demonstrating its effectiveness and generalizability.

Abstract: Learning 3D shape representation with dense correspondence for deformable objects is a fundamental problem in computer vision. Existing approaches often need additional annotations of specific semantic domain, e.g., skeleton poses for human bodies or animals, which require extra annotation effort and suffer from error accumulation, and they are limited to specific domain. In this paper, we propose a novel self-supervised approach to learn neural implicit shape representation for deformable objects, which can represent shapes with a template shape and dense correspondence in 3D. Our method does not require the priors of skeleton and skinning weight, and only requires a collection of shapes represented in signed distance fields. To handle the large deformation, we constrain the learned template shape in the same latent space with the training shapes, design a new formulation of local rigid constraint that enforces rigid transformation in local region and addresses local reflection issue, and present a new hierarchical rigid constraint to reduce the ambiguity due to the joint learning of template shape and correspondences. Extensive experiments show that our model can represent shapes with large deformations. We also show that our shape representation can support two typical applications, such as texture transfer and shape editing, with competitive performance. The code and models are available at https://iscas3dv.github.io/deformshape

Abstract: Recent advances in semi-supervised semantic segmentation have been heavily reliant on pseudo labeling to compensate for limited labeled data, disregarding the valuable relational knowledge among semantic concepts. To bridge this gap, we devise LogicDiag, a brand new neural-logic semi-supervised learning framework. Our key insight is that conflicts within pseudo labels, identified through symbolic knowledge, can serve as strong yet commonly ignored learning signals. LogicDiag resolves such conflicts via reasoning with logic-induced diagnoses, enabling the recovery of (potentially) erroneous pseudo labels, ultimately alleviating the notorious error accumulation problem. We showcase the practical application of LogicDiag in the data-hungry segmentation scenario, where we formalize the structured abstraction of semantic concepts as a set of logic rules. Extensive experiments on three standard semi-supervised semantic segmentation benchmarks demonstrate the effectiveness and generality of LogicDiag. Moreover, LogicDiag highlights the promising opportunities arising from the systematic integration of symbolic reasoning into the prevalent statistical, neural learning approaches.

Abstract: Pose based video sychronization can have applications in multiple domains such as gameplay performance evaluation, choreography or guiding athletes. The subject's actions could be compared and evaluated against those performed by professionals side by side. In this paper, we propose an end to end pipeline for synchronizing videos based on pose. The first step crops the region where the person present in the image followed by pose detection on the cropped image. This is followed by application of Dynamic Time Warping(DTW) on angle/ distance measures between the pose keypoints leading to a scale and shift invariant pose matching pipeline.

Abstract: Automatic medical report generation (MRG) is of great research value as it has the potential to relieve radiologists from the heavy burden of report writing. Despite recent advancements, accurate MRG remains challenging due to the need for precise clinical understanding and the identification of clinical findings. Moreover, the imbalanced distribution of diseases makes the challenge even more pronounced, as rare diseases are underrepresented in training data, making their diagnostic performance unreliable. To address these challenges, we propose diagnosis-driven prompts for medical report generation (PromptMRG), a novel framework that aims to improve the diagnostic accuracy of MRG with the guidance of diagnosis-aware prompts. Specifically, PromptMRG is based on encoder-decoder architecture with an extra disease classification branch. When generating reports, the diagnostic results from the classification branch are converted into token prompts to explicitly guide the generation process. To further improve the diagnostic accuracy, we design cross-modal feature enhancement, which retrieves similar reports from the database to assist the diagnosis of a query image by leveraging the knowledge from a pre-trained CLIP. Moreover, the disease imbalanced issue is addressed by applying an adaptive logit-adjusted loss to the classification branch based on the individual learning status of each disease, which overcomes the barrier of text decoder's inability to manipulate disease distributions. Experiments on two MRG benchmarks show the effectiveness of the proposed method, where it obtains state-of-the-art clinical efficacy performance on both datasets.

Abstract: Diffusion models have exhibited promising progress in video generation. However, they often struggle to retain consistent details within local regions across frames. One underlying cause is that traditional diffusion models approximate Gaussian noise distribution by utilizing predictive noise, without fully accounting for the impact of inherent information within the input itself. Additionally, these models emphasize the distinction between predictions and references, neglecting information intrinsic to the videos. To address this limitation, inspired by the self-attention mechanism, we propose a novel text-to-video (T2V) generation network structure based on diffusion models, dubbed Additional Perturbation for Latent noise with Adversarial training (APLA). Our approach only necessitates a single video as input and builds upon pre-trained stable diffusion networks. Notably, we introduce an additional compact network, known as the Video Generation Transformer (VGT). This auxiliary component is designed to extract perturbations from the inherent information contained within the input, thereby refining inconsistent pixels during temporal predictions. We leverage a hybrid architecture of transformers and convolutions to compensate for temporal intricacies, enhancing consistency between different frames within the video. Experiments demonstrate a noticeable improvement in the consistency of the generated videos both qualitatively and quantitatively.

Abstract: Point-supervised Temporal Action Localization (PSTAL) is an emerging research direction for label-efficient learning. However, current methods mainly focus on optimizing the network either at the snippet-level or the instance-level, neglecting the inherent reliability of point annotations at both levels. In this paper, we propose a Hierarchical Reliability Propagation (HR-Pro) framework, which consists of two reliability-aware stages: Snippet-level Discrimination Learning and Instance-level Completeness Learning, both stages explore the efficient propagation of high-confidence cues in point annotations. For snippet-level learning, we introduce an online-updated memory to store reliable snippet prototypes for each class. We then employ a Reliability-aware Attention Block to capture both intra-video and inter-video dependencies of snippets, resulting in more discriminative and robust snippet representation. For instance-level learning, we propose a point-based proposal generation approach as a means of connecting snippets and instances, which produces high-confidence proposals for further optimization at the instance level. Through multi-level reliability-aware learning, we obtain more reliable confidence scores and more accurate temporal boundaries of predicted proposals. Our HR-Pro achieves state-of-the-art performance on multiple challenging benchmarks, including an impressive average mAP of 60.3% on THUMOS14. Notably, our HR-Pro largely surpasses all previous point-supervised methods, and even outperforms several competitive fully supervised methods. Code will be available at https://github.com/pipixin321/HR-Pro.

Abstract: Weakly supervised temporal action localization (WSTAL) aims to localize actions in untrimmed videos using video-level labels. Despite recent advances, existing approaches mainly follow a localization-by-classification pipeline, generally processing each segment individually, thereby exploiting only limited contextual information. As a result, the model will lack a comprehensive understanding (e.g. appearance and temporal structure) of various action patterns, leading to ambiguity in classification learning and temporal localization. Our work addresses this from a novel perspective, by exploring and exploiting the cross-video contextual knowledge within the dataset to recover the dataset-level semantic structure of action instances via weak labels only, thereby indirectly improving the holistic understanding of fine-grained action patterns and alleviating the aforementioned ambiguities. Specifically, an end-to-end framework is proposed, including a Robust Memory-Guided Contrastive Learning (RMGCL) module and a Global Knowledge Summarization and Aggregation (GKSA) module. First, the RMGCL module explores the contrast and consistency of cross-video action features, assisting in learning more structured and compact embedding space, thus reducing ambiguity in classification learning. Further, the GKSA module is used to efficiently summarize and propagate the cross-video representative action knowledge in a learnable manner to promote holistic action patterns understanding, which in turn allows the generation of high-confidence pseudo-labels for self-learning, thus alleviating ambiguity in temporal localization. Extensive experiments on THUMOS14, ActivityNet1.3, and FineAction demonstrate that our method outperforms the state-of-the-art methods, and can be easily plugged into other WSTAL methods.

Abstract: The classification of gigapixel histopathology images with deep multiple instance learning models has become a critical task in digital pathology and precision medicine. In this work, we propose a Transformer-based multiple instance learning approach that replaces the traditional learned attention mechanism with a regional, Vision Transformer inspired self-attention mechanism. We present a method that fuses regional patch information to derive slide-level predictions and show how this regional aggregation can be stacked to hierarchically process features on different distance levels. To increase predictive accuracy, especially for datasets with small, local morphological features, we introduce a method to focus the image processing on high attention regions during inference. Our approach is able to significantly improve performance over the baseline on two histopathology datasets and points towards promising directions for further research.

Abstract: Currently, digital avatars can be created manually using human images as reference. Systems such as Bitmoji are excellent producers of detailed avatar designs, with hundreds of choices for customization. A supervised learning model could be trained to generate avatars automatically, but the hundreds of possible options create difficulty in securing non-noisy data to train a model. As a solution, we train a model to produce avatars from human images using tag-based annotations. This method provides better annotator agreement, leading to less noisy data and higher quality model predictions. Our contribution is an application of tag-based annotation to train a model for avatar face creation. We design tags for 3 different facial facial features offered by Bitmoji, and train a model using tag-based annotation to predict the nose.

Abstract: The CoachAI Badminton 2023 Track1 initiative aim to automatically detect events within badminton match videos. Detecting small objects, especially the shuttlecock, is of quite importance and demands high precision within the challenge. Such detection is crucial for tasks like hit count, hitting time, and hitting location. However, even after revising the well-regarded shuttlecock detecting model, TrackNet, our object detection models still fall short of the desired accuracy. To address this issue, we've implemented various deep learning methods to tackle the problems arising from noisy detectied data, leveraging diverse data types to improve precision. In this report, we detail the detection model modifications we've made and our approach to the 11 tasks. Notably, our system garnered a score of 0.78 out of 1.0 in the challenge.

Abstract: Assessing the robustness of deep neural networks against out-of-distribution inputs is crucial, especially in safety-critical domains like autonomous driving, but also in safety systems where malicious actors can digitally alter inputs to circumvent safety guards. However, designing effective out-of-distribution tests that encompass all possible scenarios while preserving accurate label information is a challenging task. Existing methodologies often entail a compromise between variety and constraint levels for attacks and sometimes even both. In a first step towards a more holistic robustness evaluation of image classification models, we introduce an attack method based on image solarization that is conceptually straightforward yet avoids jeopardizing the global structure of natural images independent of the intensity. Through comprehensive evaluations of multiple ImageNet models, we demonstrate the attack's capacity to degrade accuracy significantly, provided it is not integrated into the training augmentations. Interestingly, even then, no full immunity to accuracy deterioration is achieved. In other settings, the attack can often be simplified into a black-box attack with model-independent parameters. Defenses against other corruptions do not consistently extend to be effective against our specific attack. Project website: https://github.com/paulgavrikov/adversarial_solarization

Abstract: In this paper, we introduce Masked Feature Modelling (MFM), a novel approach for the unsupervised pre-training of a Graph Attention Network (GAT) block. MFM utilizes a pretrained Visual Tokenizer to reconstruct masked features of objects within a video, leveraging the MiniKinetics dataset. We then incorporate the pre-trained GAT block into a state-of-the-art bottom-up supervised video-event recognition architecture, ViGAT, to improve the model's starting point and overall accuracy. Experimental evaluations on the YLI-MED dataset demonstrate the effectiveness of MFM in improving event recognition performance.

Abstract: Accurate classification of Acute Myeloid Leukemia (AML) subtypes is crucial for clinical decision-making and patient care. In this study, we investigate the potential presence of age and sex bias in AML subtype classification using Multiple Instance Learning (MIL) architectures. To that end, we train multiple MIL models using different levels of sex imbalance in the training set and excluding certain age groups. To assess the sex bias, we evaluate the performance of the models on male and female test sets. For age bias, models are tested against underrepresented age groups in the training data. We find a significant effect of sex and age bias on the performance of the model for AML subtype classification. Specifically, we observe that females are more likely to be affected by sex imbalance dataset and certain age groups, such as patients with 72 to 86 years of age with the RUNX1::RUNX1T1 genetic subtype, are significantly affected by an age bias present in the training data. Ensuring inclusivity in the training data is thus essential for generating reliable and equitable outcomes in AML genetic subtype classification, ultimately benefiting diverse patient populations.

Abstract: Accurate classification of white blood cells in peripheral blood is essential for diagnosing hematological diseases. Due to constantly evolving clinical settings, data sources, and disease classifications, it is necessary to update machine learning classification models regularly for practical real-world use. Such models significantly benefit from sequentially learning from incoming data streams without forgetting previously acquired knowledge. However, models can suffer from catastrophic forgetting, causing a drop in performance on previous tasks when fine-tuned on new data. Here, we propose a rehearsal-based continual learning approach for class incremental and domain incremental scenarios in white blood cell classification. To choose representative samples from previous tasks, we employ exemplar set selection based on the model's predictions. This involves selecting the most confident samples and the most challenging samples identified through uncertainty estimation of the model. We thoroughly evaluated our proposed approach on three white blood cell classification datasets that differ in color, resolution, and class composition, including scenarios where new domains or new classes are introduced to the model with every task. We also test a long class incremental experiment with both new domains and new classes. Our results demonstrate that our approach outperforms established baselines in continual learning, including existing iCaRL and EWC methods for classifying white blood cells in cross-domain environments.

Abstract: Creating layouts is a fundamental step in graphic design. In this work, we propose to use text as the guidance to create graphic layouts, i.e., Text-to-Layout, aiming to lower the design barriers. Text-to-Layout is a challenging task, because it needs to consider the implicit, combined, and incomplete layout constraints from text, each of which has not been studied in previous work. To address this, we present a two-stage approach, named parse-then-place. The approach introduces an intermediate representation (IR) between text and layout to represent diverse layout constraints. With IR, Text-to-Layout is decomposed into a parse stage and a place stage. The parse stage takes a textual description as input and generates an IR, in which the implicit constraints from the text are transformed into explicit ones. The place stage generates layouts based on the IR. To model combined and incomplete constraints, we use a Transformer-based layout generation model and carefully design a way to represent constraints and layouts as sequences. Besides, we adopt the pretrain-then-finetune strategy to boost the performance of the layout generation model with large-scale unlabeled layouts. To evaluate our approach, we construct two Text-to-Layout datasets and conduct experiments on them. Quantitative results, qualitative analysis, and user studies demonstrate the effectiveness of our approach.

Abstract: In this work, we construct a large-scale dataset for Ground-to-Aerial Person Search, named G2APS, which contains 31,770 images of 260,559 annotated bounding boxes for 2,644 identities appearing in both of the UAVs and ground surveillance cameras. To our knowledge, this is the first dataset for cross-platform intelligent surveillance applications, where the UAVs could work as a powerful complement for the ground surveillance cameras. To more realistically simulate the actual cross-platform Ground-to-Aerial surveillance scenarios, the surveillance cameras are fixed about 2 meters above the ground, while the UAVs capture videos of persons at different location, with a variety of view-angles, flight attitudes and flight modes. Therefore, the dataset has the following unique characteristics: 1) drastic view-angle changes between query and gallery person images from cross-platform cameras; 2) diverse resolutions, poses and views of the person images under 9 rich real-world scenarios. On basis of the G2APS benchmark dataset, we demonstrate detailed analysis about current two-step and end-to-end person search methods, and further propose a simple yet effective knowledge distillation scheme on the head of the ReID network, which achieves state-of-the-art performances on both of the G2APS and the previous two public person search datasets, i.e., PRW and CUHK-SYSU. The dataset and source code available on \url{https://github.com/yqc123456/HKD_for_person_search}.

Abstract: The integration of visual encoders and large language models (LLMs) has driven recent progress in multimodal large language models (MLLMs). However, the scarcity of high-quality instruction-tuning data for vision-language tasks remains a challenge. The current leading paradigm, such as LLaVA, relies on language-only GPT-4 to generate data, which requires pre-annotated image captions and detection bounding boxes, suffering from understanding image details. A practical solution to this problem would be to utilize the available multimodal large language models (MLLMs) to generate instruction data for vision-language tasks. However, it's worth noting that the currently accessible MLLMs are not as powerful as their LLM counterparts, as they tend to produce inadequate responses and generate false information. As a solution for addressing the current issue, this paper proposes the Visual Instruction Generation and Correction (VIGC) framework that enables multimodal large language models to generate instruction-tuning data and progressively enhance its quality on-the-fly. Specifically, Visual Instruction Generation (VIG) guides the vision-language model to generate diverse instruction-tuning data. To ensure generation quality, Visual Instruction Correction (VIC) adopts an iterative update mechanism to correct any inaccuracies in data produced by VIG, effectively reducing the risk of hallucination. Leveraging the diverse, high-quality data generated by VIGC, we finetune mainstream models and validate data quality based on various evaluations. Experimental results demonstrate that VIGC not only compensates for the shortcomings of language-only data generation methods, but also effectively enhances the benchmark performance. The models, datasets, and code will be made publicly available.

Abstract: In recent years, computer-aided diagnosis systems have shown great potential in assisting radiologists with accurate and efficient medical image analysis. This paper presents a novel approach for bone pathology localization and classification in wrist X-ray images using a combination of YOLO (You Only Look Once) and the Shifted Window Transformer (Swin) with a newly proposed block. The proposed methodology addresses two critical challenges in wrist X-ray analysis: accurate localization of bone pathologies and precise classification of abnormalities. The YOLO framework is employed to detect and localize bone pathologies, leveraging its real-time object detection capabilities. Additionally, the Swin, a transformer-based module, is utilized to extract contextual information from the localized regions of interest (ROIs) for accurate classification.

Abstract: The hypothalamus plays a crucial role in the regulation of a broad range of physiological, behavioural, and cognitive functions. However, despite its importance, only a few small-scale neuroimaging studies have investigated its substructures, likely due to the lack of fully automated segmentation tools to address scalability and reproducibility issues of manual segmentation. While the only previous attempt to automatically sub-segment the hypothalamus with a neural network showed promise for 1.0 mm isotropic T1-weighted (T1w) MRI, there is a need for an automated tool to sub-segment also high-resolutional (HiRes) MR scans, as they are becoming widely available, and include structural detail also from multi-modal MRI. We, therefore, introduce a novel, fast, and fully automated deep learning method named HypVINN for sub-segmentation of the hypothalamus and adjacent structures on 0.8 mm isotropic T1w and T2w brain MR images that is robust to missing modalities. We extensively validate our model with respect to segmentation accuracy, generalizability, in-session test-retest reliability, and sensitivity to replicate hypothalamic volume effects (e.g. sex-differences). The proposed method exhibits high segmentation performance both for standalone T1w images as well as for T1w/T2w image pairs. Even with the additional capability to accept flexible inputs, our model matches or exceeds the performance of state-of-the-art methods with fixed inputs. We, further, demonstrate the generalizability of our method in experiments with 1.0 mm MR scans from both the Rhineland Study and the UK Biobank. Finally, HypVINN can perform the segmentation in less than a minute (GPU) and will be available in the open source FastSurfer neuroimaging software suite, offering a validated, efficient, and scalable solution for evaluating imaging-derived phenotypes of the hypothalamus.

Abstract: Convolutional neural networks excel in histopathological image classification, yet their pixel-level focus hampers explainability. Conversely, emerging graph convolutional networks spotlight cell-level features and medical implications. However, limited by their shallowness and suboptimal use of high-dimensional pixel data, GCNs underperform in multi-class histopathological image classification. To make full use of pixel-level and cell-level features dynamically, we propose an asymmetric co-training framework combining a deep graph convolutional network and a convolutional neural network for multi-class histopathological image classification. To improve the explainability of the entire framework by embedding morphological and topological distribution of cells, we build a 14-layer deep graph convolutional network to handle cell graph data. For the further utilization and dynamic interactions between pixel-level and cell-level information, we also design a co-training strategy to integrate the two asymmetric branches. Notably, we collect a private clinically acquired dataset termed LUAD7C, including seven subtypes of lung adenocarcinoma, which is rare and more challenging. We evaluated our approach on the private LUAD7C and public colorectal cancer datasets, showcasing its superior performance, explainability, and generalizability in multi-class histopathological image classification.

Abstract: Underwater object detection suffers from low detection performance because the distance and wavelength dependent imaging process yield evident image quality degradations such as haze-like effects, low visibility, and color distortions. Therefore, we commit to resolving the issue of underwater object detection with compounded environmental degradations. Typical approaches attempt to develop sophisticated deep architecture to generate high-quality images or features. However, these methods are only work for limited ranges because imaging factors are either unstable, too sensitive, or compounded. Unlike these approaches catering for high-quality images or features, this paper seeks transferable prior knowledge from detector-friendly images. The prior guides detectors removing degradations that interfere with detection. It is based on statistical observations that, the heavily degraded regions of detector-friendly (DFUI) and underwater images have evident feature distribution gaps while the lightly degraded regions of them overlap each other. Therefore, we propose a residual feature transference module (RFTM) to learn a mapping between deep representations of the heavily degraded patches of DFUI- and underwater- images, and make the mapping as a heavily degraded prior (HDP) for underwater detection. Since the statistical properties are independent to image content, HDP can be learned without the supervision of semantic labels and plugged into popular CNNbased feature extraction networks to improve their performance on underwater object detection. Without bells and whistles, evaluations on URPC2020 and UODD show that our methods outperform CNN-based detectors by a large margin. Our method with higher speeds and less parameters still performs better than transformer-based detectors. Our code and DFUI dataset can be found in https://github.com/xiaoDetection/Learning-Heavily-Degraed-Prior.

Abstract: In this work, we address the task of few-shot part segmentation, which aims to segment the different parts of an unseen object using very few labeled examples. It is found that leveraging the textual space of a powerful pre-trained image-language model (such as CLIP) can be beneficial in learning visual features. Therefore, we develop a novel method termed PartSeg for few-shot part segmentation based on multimodal learning. Specifically, we design a part-aware prompt learning method to generate part-specific prompts that enable the CLIP model to better understand the concept of ``part'' and fully utilize its textual space. Furthermore, since the concept of the same part under different object categories is general, we establish relationships between these parts during the prompt learning process. We conduct extensive experiments on the PartImageNet and Pascal$\_$Part datasets, and the experimental results demonstrated that our proposed method achieves state-of-the-art performance.

Abstract: The diversity in length constitutes a significant characteristic of text. Due to the long-tail distribution of text lengths, most existing methods for scene text recognition (STR) only work well on short or seen-length text, lacking the capability of recognizing longer text or performing length extrapolation. This is a crucial issue, since the lengths of the text to be recognized are usually not given in advance in real-world applications, but it has not been adequately investigated in previous works. Therefore, we propose in this paper a method called Length-Insensitive Scene TExt Recognizer (LISTER), which remedies the limitation regarding the robustness to various text lengths. Specifically, a Neighbor Decoder is proposed to obtain accurate character attention maps with the assistance of a novel neighbor matrix regardless of the text lengths. Besides, a Feature Enhancement Module is devised to model the long-range dependency with low computation cost, which is able to perform iterations with the neighbor decoder to enhance the feature map progressively. To the best of our knowledge, we are the first to achieve effective length-insensitive scene text recognition. Extensive experiments demonstrate that the proposed LISTER algorithm exhibits obvious superiority on long text recognition and the ability for length extrapolation, while comparing favourably with the previous state-of-the-art methods on standard benchmarks for STR (mainly short text).

Abstract: Prior work in 3D object detection evaluates models using offline metrics like average precision since closed-loop online evaluation on the downstream driving task is costly. However, it is unclear how indicative offline results are of driving performance. In this work, we perform the first empirical evaluation measuring how predictive different detection metrics are of driving performance when detectors are integrated into a full self-driving stack. We conduct extensive experiments on urban driving in the CARLA simulator using 16 object detection models. We find that the nuScenes Detection Score has a higher correlation to driving performance than the widely used average precision metric. In addition, our results call for caution on the exclusive reliance on the emerging class of `planner-centric' metrics.

Abstract: Surgical context inference has recently garnered significant attention in robot-assisted surgery as it can facilitate workflow analysis, skill assessment, and error detection. However, runtime context inference is challenging since it requires timely and accurate detection of the interactions among the tools and objects in the surgical scene based on the segmentation of video data. On the other hand, existing state-of-the-art video segmentation methods are often biased against infrequent classes and fail to provide temporal consistency for segmented masks. This can negatively impact the context inference and accurate detection of critical states. In this study, we propose a solution to these challenges using a Space Time Correspondence Network (STCN). STCN is a memory network that performs binary segmentation and minimizes the effects of class imbalance. The use of a memory bank in STCN allows for the utilization of past image and segmentation information, thereby ensuring consistency of the masks. Our experiments using the publicly available JIGSAWS dataset demonstrate that STCN achieves superior segmentation performance for objects that are difficult to segment, such as needle and thread, and improves context inference compared to the state-of-the-art. We also demonstrate that segmentation and context inference can be performed at runtime without compromising performance.

Abstract: The portrait matting task aims to extract an alpha matte with complete semantics and finely-detailed contours. In comparison to CNN-based approaches, transformers with self-attention allow a larger receptive field, enabling it to better capture long-range dependencies and low-frequency semantic information of a portrait. However, the recent research shows that self-attention mechanism struggle with modeling high-frequency information and capturing fine contour details, which can lead to bias while predicting the portrait's contours. To address the problem, we propose EFormer to enhance the model's attention towards semantic and contour features. Especially the latter, which is surrounded by a large amount of high-frequency details. We build a semantic and contour detector (SCD) to accurately capture the distribution of semantic and contour features. And we further design contour-edge extraction branch and semantic extraction branch for refining contour features and complete semantic information. Finally, we fuse the two kinds of features and leverage the segmentation head to generate the predicted portrait matte. Remarkably, EFormer is an end-to-end trimap-free method and boasts a simple structure. Experiments conducted on VideoMatte240K-JPEGSD and AIM datasets demonstrate that EFormer outperforms previous portrait matte methods.

Abstract: Through our respiratory system, many viruses and diseases frequently spread and pass from one person to another. Covid-19 served as an example of how crucial it is to track down and cut back on contacts to stop its spread. There is a clear gap in finding automatic methods that can detect hand-to-face contact in complex urban scenes or indoors. In this paper, we introduce a computer vision framework, called FaceTouch, based on deep learning. It comprises deep sub-models to detect humans and analyse their actions. FaceTouch seeks to detect hand-to-face touches in the wild, such as through video chats, bus footage, or CCTV feeds. Despite partial occlusion of faces, the introduced system learns to detect face touches from the RGB representation of a given scene by utilising the representation of the body gestures such as arm movement. This has been demonstrated to be useful in complex urban scenarios beyond simply identifying hand movement and its closeness to faces. Relying on Supervised Contrastive Learning, the introduced model is trained on our collected dataset, given the absence of other benchmark datasets. The framework shows a strong validation in unseen datasets which opens the door for potential deployment.

Abstract: Robust obstacle avoidance is one of the critical steps for successful goal-driven indoor navigation tasks.Due to the obstacle missing in the visual image and the possible missed detection issue, visual image-based obstacle avoidance techniques still suffer from unsatisfactory robustness. To mitigate it, in this paper, we propose a novel implicit obstacle map-driven indoor navigation framework for robust obstacle avoidance, where an implicit obstacle map is learned based on the historical trial-and-error experience rather than the visual image. In order to further improve the navigation efficiency, a non-local target memory aggregation module is designed to leverage a non-local network to model the intrinsic relationship between the target semantic and the target orientation clues during the navigation process so as to mine the most target-correlated object clues for the navigation decision. Extensive experimental results on AI2-Thor and RoboTHOR benchmarks verify the excellent obstacle avoidance and navigation efficiency of our proposed method. The core source code is available at https://github.com/xwaiyy123/object-navigation.

Abstract: Multi-modal fusion is increasingly being used for autonomous driving tasks, as images from different modalities provide unique information for feature extraction. However, the existing two-stream networks are only fused at a specific network layer, which requires a lot of manual attempts to set up. As the CNN goes deeper, the two modal features become more and more advanced and abstract, and the fusion occurs at the feature level with a large gap, which can easily hurt the performance. In this study, we propose a novel fusion architecture called skip-cross networks (SkipcrossNets), which combines adaptively LiDAR point clouds and camera images without being bound to a certain fusion epoch. Specifically, skip-cross connects each layer to each layer in a feed-forward manner, and for each layer, the feature maps of all previous layers are used as input and its own feature maps are used as input to all subsequent layers for the other modality, enhancing feature propagation and multi-modal features fusion. This strategy facilitates selection of the most similar feature layers from two data pipelines, providing a complementary effect for sparse point cloud features during fusion processes. The network is also divided into several blocks to reduce the complexity of feature fusion and the number of model parameters. The advantages of skip-cross fusion were demonstrated through application to the KITTI and A2D2 datasets, achieving a MaxF score of 96.85% on KITTI and an F1 score of 84.84% on A2D2. The model parameters required only 2.33 MB of memory at a speed of 68.24 FPS, which could be viable for mobile terminals and embedded devices.

Abstract: We target cross-domain face reenactment in this paper, i.e., driving a cartoon image with the video of a real person and vice versa. Recently, many works have focused on one-shot talking face generation to drive a portrait with a real video, i.e., within-domain reenactment. Straightforwardly applying those methods to cross-domain animation will cause inaccurate expression transfer, blur effects, and even apparent artifacts due to the domain shift between cartoon and real faces. Only a few works attempt to settle cross-domain face reenactment. The most related work AnimeCeleb requires constructing a dataset with pose vector and cartoon image pairs by animating 3D characters, which makes it inapplicable anymore if no paired data is available. In this paper, we propose a novel method for cross-domain reenactment without paired data. Specifically, we propose a transformer-based framework to align the motions from different domains into a common latent space where motion transfer is conducted via latent code addition. Two domain-specific motion encoders and two learnable motion base memories are used to capture domain properties. A source query transformer and a driving one are exploited to project domain-specific motion to the canonical space. The edited motion is projected back to the domain of the source with a transformer. Moreover, since no paired data is provided, we propose a novel cross-domain training scheme using data from two domains with the designed analogy constraint. Besides, we contribute a cartoon dataset in Disney style. Extensive evaluations demonstrate the superiority of our method over competing methods.

Abstract: LiDAR-based place recognition plays a crucial role in Simultaneous Localization and Mapping (SLAM) and LiDAR localization. Despite the emergence of various deep learning-based and hand-crafting-based methods, rotation-induced place recognition failure remains a critical challenge. Existing studies address this limitation through specific training strategies or network structures. However, the former does not produce satisfactory results, while the latter focuses mainly on the reduced problem of SO(2) rotation invariance. Methods targeting SO(3) rotation invariance suffer from limitations in discrimination capability. In this paper, we propose a new method that employs Vector Neurons Network (VNN) to achieve SO(3) rotation invariance. We first extract rotation-equivariant features from neighboring points and map low-dimensional features to a high-dimensional space through VNN. Afterwards, we calculate the Euclidean and Cosine distance in the rotation-equivariant feature space as rotation-invariant feature descriptors. Finally, we aggregate the features using GeM pooling to obtain global descriptors. To address the significant information loss when formulating rotation-invariant descriptors, we propose computing distances between features at different layers within the Euclidean space neighborhood. This greatly improves the discriminability of the point cloud descriptors while ensuring computational efficiency. Experimental results on public datasets show that our approach significantly outperforms other baseline methods implementing rotation invariance, while achieving comparable results with current state-of-the-art place recognition methods that do not consider rotation issues.

Abstract: For the convolutional neural network (CNN) used for pattern classification, the training loss function is usually applied to the final output of the network, except for some regularization constraints on the network parameters. However, with the increasing of the number of network layers, the influence of the loss function on the network front layers gradually decreases, and the network parameters tend to fall into local optimization. At the same time, it is found that the trained network has significant information redundancy at all stages of features, which reduces the effectiveness of feature mapping at all stages and is not conducive to the change of the subsequent parameters of the network in the direction of optimality. Therefore, it is possible to obtain a more optimized solution of the network and further improve the classification accuracy of the network by designing a loss function for restraining the front stage features and eliminating the information redundancy of the front stage features .For CNN, this article proposes a multi-stage feature decorrelation loss (MFD Loss), which refines effective features and eliminates information redundancy by constraining the correlation of features at all stages. Considering that there are many layers in CNN, through experimental comparison and analysis, MFD Loss acts on multiple front layers of CNN, constrains the output features of each layer and each channel, and performs supervision training jointly with classification loss function during network training. Compared with the single Softmax Loss supervised learning, the experiments on several commonly used datasets on several typical CNNs prove that the classification performance of Softmax Loss+MFD Loss is significantly better. Meanwhile, the comparison experiments before and after the combination of MFD Loss and some other typical loss functions verify its good universality.

Abstract: Semantic segmentation is a computer vision task that associates a label with each pixel in an image. Modern approaches tend to introduce class embeddings into semantic segmentation for deeply utilizing category semantics, and regard supervised class masks as final predictions. In this paper, we explore the mechanism of class embeddings and have an insight that more explicit and meaningful class embeddings can be generated based on class masks purposely. Following this observation, we propose ECENet, a new segmentation paradigm, in which class embeddings are obtained and enhanced explicitly during interacting with multi-stage image features. Based on this, we revisit the traditional decoding process and explore inverted information flow between segmentation masks and class embeddings. Furthermore, to ensure the discriminability and informativity of features from backbone, we propose a Feature Reconstruction module, which combines intrinsic and diverse branches together to ensure the concurrence of diversity and redundancy in features. Experiments show that our ECENet outperforms its counterparts on the ADE20K dataset with much less computational cost and achieves new state-of-the-art results on PASCAL-Context dataset. The code will be released at https://gitee.com/mindspore/models and https://github.com/Carol-lyh/ECENet.

Abstract: This paper highlights the need to bring document classification benchmarking closer to real-world applications, both in the nature of data tested ($X$: multi-channel, multi-paged, multi-industry; $Y$: class distributions and label set variety) and in classification tasks considered ($f$: multi-page document, page stream, and document bundle classification, ...). We identify the lack of public multi-page document classification datasets, formalize different classification tasks arising in application scenarios, and motivate the value of targeting efficient multi-page document representations. An experimental study on proposed multi-page document classification datasets demonstrates that current benchmarks have become irrelevant and need to be updated to evaluate complete documents, as they naturally occur in practice. This reality check also calls for more mature evaluation methodologies, covering calibration evaluation, inference complexity (time-memory), and a range of realistic distribution shifts (e.g., born-digital vs. scanning noise, shifting page order). Our study ends on a hopeful note by recommending concrete avenues for future improvements.}

Abstract: Low-light images, characterized by inadequate illumination, pose challenges of diminished clarity, muted colors, and reduced details. Low-light image enhancement, an essential task in computer vision, aims to rectify these issues by improving brightness, contrast, and overall perceptual quality, thereby facilitating accurate analysis and interpretation. This paper introduces the Convolutional Dense Attention-guided Network (CDAN), a novel solution for enhancing low-light images. CDAN integrates an autoencoder-based architecture with convolutional and dense blocks, complemented by an attention mechanism and skip connections. This architecture ensures efficient information propagation and feature learning. Furthermore, a dedicated post-processing phase refines color balance and contrast. Our approach demonstrates notable progress compared to state-of-the-art results in low-light image enhancement, showcasing its robustness across a wide range of challenging scenarios. Our model performs remarkably on benchmark datasets, effectively mitigating under-exposure and proficiently restoring textures and colors in diverse low-light scenarios. This achievement underscores CDAN's potential for diverse computer vision tasks, notably enabling robust object detection and recognition in challenging low-light conditions.

Abstract: We propose Subject-Conditional Relation Detection SCoRD, where conditioned on an input subject, the goal is to predict all its relations to other objects in a scene along with their locations. Based on the Open Images dataset, we propose a challenging OIv6-SCoRD benchmark such that the training and testing splits have a distribution shift in terms of the occurrence statistics of $\langle$subject, relation, object$\rangle$ triplets. To solve this problem, we propose an auto-regressive model that given a subject, it predicts its relations, objects, and object locations by casting this output as a sequence of tokens. First, we show that previous scene-graph prediction methods fail to produce as exhaustive an enumeration of relation-object pairs when conditioned on a subject on this benchmark. Particularly, we obtain a recall@3 of 83.8% for our relation-object predictions compared to the 49.75% obtained by a recent scene graph detector. Then, we show improved generalization on both relation-object and object-box predictions by leveraging during training relation-object pairs obtained automatically from textual captions and for which no object-box annotations are available. Particularly, for $\langle$subject, relation, object$\rangle$ triplets for which no object locations are available during training, we are able to obtain a recall@3 of 42.59% for relation-object pairs and 32.27% for their box locations.

Abstract: In recent years, the effective and safe collaboration between humans and machines has gained significant importance, particularly in the Industry 4.0 scenario. A critical prerequisite for realizing this collaborative paradigm is precisely understanding the robot's 3D pose within its environment. Therefore, in this paper, we introduce a novel vision-based system leveraging depth data to accurately establish the 3D locations of robotic joints. Specifically, we prove the ability of the proposed system to enhance its current pose estimation accuracy by jointly learning to forecast future poses. Indeed, we introduce the concept of Pose Nowcasting, denoting the capability of a system to exploit the learned knowledge of the future to improve the estimation of the present. The experimental evaluation is conducted on two different datasets, providing state-of-the-art and real-time performance and confirming the validity of the proposed method on both the robotic and human scenarios.

Abstract: The emergence of vision-language models (VLMs), such as CLIP, has spurred a significant research effort towards their application for downstream supervised learning tasks. Although some previous studies have explored the unsupervised fine-tuning of CLIP, they often rely on prior knowledge in the form of class names associated with ground truth labels. In this paper, we delve into a realistic unsupervised fine-tuning scenario by assuming that the unlabeled data might contain out-of-distribution samples from unknown classes. Furthermore, we emphasize the importance of simultaneously enhancing out-of-distribution detection capabilities alongside the recognition of instances associated with predefined class labels. To tackle this problem, we present a simple, efficient, and effective fine-tuning approach called Universal Entropy Optimization (UEO). UEO leverages sample-level confidence to approximately minimize the conditional entropy of confident instances and maximize the marginal entropy of less confident instances. Apart from optimizing the textual prompts, UEO also incorporates optimization of channel-wise affine transformations within the visual branch of CLIP. Through extensive experiments conducted across 15 domains and 4 different types of prior knowledge, we demonstrate that UEO surpasses baseline methods in terms of both generalization and out-of-distribution detection.

Abstract: Image segmentation and depth estimation are crucial tasks in computer vision, especially in autonomous driving scenarios. Although these tasks are typically addressed separately, we propose an innovative approach to combine them in our novel deep learning network, Panoptic-DepthLab. By incorporating an additional depth estimation branch into the segmentation network, it can predict the depth of each instance segment. Evaluating on Cityscape dataset, we demonstrate the effectiveness of our method in achieving high-quality segmentation results with depth and visualize it with a color map. Our proposed method demonstrates a new possibility of combining different tasks and networks to generate a more comprehensive image recognition result to facilitate the safety of autonomous driving vehicles.

Abstract: Monocular 3D object detection is a crucial and challenging task for autonomous driving vehicle, while it uses only a single camera image to infer 3D objects in the scene. To address the difficulty of predicting depth using only pictorial clue, we propose a novel perspective-aware convolutional layer that captures long-range dependencies in images. By enforcing convolutional kernels to extract features along the depth axis of every image pixel, we incorporates perspective information into network architecture. We integrate our perspective-aware convolutional layer into a 3D object detector and demonstrate improved performance on the KITTI3D dataset, achieving a 23.9\% average precision in the easy benchmark. These results underscore the importance of modeling scene clues for accurate depth inference and highlight the benefits of incorporating scene structure in network design. Our perspective-aware convolutional layer has the potential to enhance object detection accuracy by providing more precise and context-aware feature extraction.

Abstract: Vision-Language Pre-training (VLP) shows remarkable progress with the assistance of extremely heavy parameters, which challenges deployment in real applications. Knowledge distillation is well recognized as the essential procedure in model compression. However, existing knowledge distillation techniques lack an in-depth investigation and analysis of VLP, and practical guidelines for VLP-oriented distillation are still not yet explored. In this paper, we present DLIP, a simple yet efficient Distilling Language-Image Pre-training framework, through which we investigate how to distill a light VLP model. Specifically, we dissect the model distillation from multiple dimensions, such as the architecture characteristics of different modules and the information transfer of different modalities. We conduct comprehensive experiments and provide insights on distilling a light but performant VLP model. Experimental results reveal that DLIP can achieve a state-of-the-art accuracy/efficiency trade-off across diverse cross-modal tasks, e.g., image-text retrieval, image captioning and visual question answering. For example, DLIP compresses BLIP by 1.9x, from 213M to 108M parameters, while achieving comparable or better performance. Furthermore, DLIP succeeds in retaining more than 95% of the performance with 22.4% parameters and 24.8% FLOPs compared to the teacher model and accelerates inference speed by 2.7x.

Abstract: Large-scale pre-trained Vision Language Models (VLMs) have proven effective for zero-shot classification. Despite the success, most traditional VLMs-based methods are restricted by the assumption of partial source supervision or ideal vocabularies, which rarely satisfy the open-world scenario. In this paper, we aim at a more challenging setting, Realistic Zero-Shot Classification, which assumes no annotation but instead a broad vocabulary. To address this challenge, we propose the Self Structural Semantic Alignment (S^3A) framework, which extracts the structural semantic information from unlabeled data while simultaneously self-learning. Our S^3A framework adopts a unique Cluster-Vote-Prompt-Realign (CVPR) algorithm, which iteratively groups unlabeled data to derive structural semantics for pseudo-supervision. Our CVPR process includes iterative clustering on images, voting within each cluster to identify initial class candidates from the vocabulary, generating discriminative prompts with large language models to discern confusing candidates, and realigning images and the vocabulary as structural semantic alignment. Finally, we propose to self-learn the CLIP image encoder with both individual and structural semantic alignment through a teacher-student learning strategy. Our comprehensive experiments across various generic and fine-grained benchmarks demonstrate that the S^3A method offers substantial improvements over existing VLMs-based approaches, achieving a more than 15% accuracy improvement over CLIP on average. Our codes, models, and prompts are publicly released at https://github.com/sheng-eatamath/S3A.

Abstract: To reduce the reliance on large-scale datasets, recent works in 3D segmentation resort to few-shot learning. Current 3D few-shot semantic segmentation methods first pre-train the models on `seen' classes, and then evaluate their generalization performance on `unseen' classes. However, the prior pre-training stage not only introduces excessive time overhead, but also incurs a significant domain gap on `unseen' classes. To tackle these issues, we propose an efficient Training-free Few-shot 3D Segmentation netwrok, TFS3D, and a further training-based variant, TFS3D-T. Without any learnable parameters, TFS3D extracts dense representations by trigonometric positional encodings, and achieves comparable performance to previous training-based methods. Due to the elimination of pre-training, TFS3D can alleviate the domain gap issue and save a substantial amount of time. Building upon TFS3D, TFS3D-T only requires to train a lightweight query-support transferring attention (QUEST), which enhances the interaction between the few-shot query and support data. Experiments demonstrate TFS3D-T improves previous state-of-the-art methods by +6.93% and +17.96% mIoU respectively on S3DIS and ScanNet, while reducing the training time by -90%, indicating superior effectiveness and efficiency.

Abstract: Several recent works have directly extended the image masked autoencoder (MAE) with random masking into video domain, achieving promising results. However, unlike images, both spatial and temporal information are important for video understanding. This suggests that the random masking strategy that is inherited from the image MAE is less effective for video MAE. This motivates the design of a novel masking algorithm that can more efficiently make use of video saliency. Specifically, we propose a motion-guided masking algorithm (MGM) which leverages motion vectors to guide the position of each mask over time. Crucially, these motion-based correspondences can be directly obtained from information stored in the compressed format of the video, which makes our method efficient and scalable. On two challenging large-scale video benchmarks (Kinetics-400 and Something-Something V2), we equip video MAE with our MGM and achieve up to +$1.3\%$ improvement compared to previous state-of-the-art methods. Additionally, our MGM achieves equivalent performance to previous video MAE using up to $66\%$ fewer training epochs. Lastly, we show that MGM generalizes better to downstream transfer learning and domain adaptation tasks on the UCF101, HMDB51, and Diving48 datasets, achieving up to +$4.9\%$ improvement compared to baseline methods.

Abstract: Despite tremendous advancements in bird's-eye view (BEV) perception, existing models fall short in generating realistic and coherent semantic map layouts, and they fail to account for uncertainties arising from partial sensor information (such as occlusion or limited coverage). In this work, we introduce MapPrior, a novel BEV perception framework that combines a traditional discriminative BEV perception model with a learned generative model for semantic map layouts. Our MapPrior delivers predictions with better accuracy, realism, and uncertainty awareness. We evaluate our model on the large-scale nuScenes benchmark. At the time of submission, MapPrior outperforms the strongest competing method, with significantly improved MMD and ECE scores in camera- and LiDAR-based BEV perception.

Abstract: Existing text-to-image diffusion models struggle to synthesize realistic images given dense captions, where each text prompt provides a detailed description for a specific image region. To address this, we propose DenseDiffusion, a training-free method that adapts a pre-trained text-to-image model to handle such dense captions while offering control over the scene layout. We first analyze the relationship between generated images' layouts and the pre-trained model's intermediate attention maps. Next, we develop an attention modulation method that guides objects to appear in specific regions according to layout guidance. Without requiring additional fine-tuning or datasets, we improve image generation performance given dense captions regarding both automatic and human evaluation scores. In addition, we achieve similar-quality visual results with models specifically trained with layout conditions.

Abstract: The regression of 3D Human Pose and Shape (HPS) from an image is becoming increasingly accurate. This makes the results useful for downstream tasks like human action recognition or 3D graphics. Yet, no regressor is perfect, and accuracy can be affected by ambiguous image evidence or by poses and appearance that are unseen during training. Most current HPS regressors, however, do not report the confidence of their outputs, meaning that downstream tasks cannot differentiate accurate estimates from inaccurate ones. To address this, we develop POCO, a novel framework for training HPS regressors to estimate not only a 3D human body, but also their confidence, in a single feed-forward pass. Specifically, POCO estimates both the 3D body pose and a per-sample variance. The key idea is to introduce a Dual Conditioning Strategy (DCS) for regressing uncertainty that is highly correlated to pose reconstruction quality. The POCO framework can be applied to any HPS regressor and here we evaluate it by modifying HMR, PARE, and CLIFF. In all cases, training the network to reason about uncertainty helps it learn to more accurately estimate 3D pose. While this was not our goal, the improvement is modest but consistent. Our main motivation is to provide uncertainty estimates for downstream tasks; we demonstrate this in two ways: (1) We use the confidence estimates to bootstrap HPS training. Given unlabelled image data, we take the confident estimates of a POCO-trained regressor as pseudo ground truth. Retraining with this automatically-curated data improves accuracy. (2) We exploit uncertainty in video pose estimation by automatically identifying uncertain frames (e.g. due to occlusion) and inpainting these from confident frames. Code and models will be available for research at https://poco.is.tue.mpg.de.

Abstract: We introduce the Qwen-VL series, a set of large-scale vision-language models designed to perceive and understand both text and images. Comprising Qwen-VL and Qwen-VL-Chat, these models exhibit remarkable performance in tasks like image captioning, question answering, visual localization, and flexible interaction. The evaluation covers a wide range of tasks including zero-shot captioning, visual or document visual question answering, and grounding. We demonstrate the Qwen-VL outperforms existing Large Vision Language Models (LVLMs). We present their architecture, training, capabilities, and performance, highlighting their contributions to advancing multimodal artificial intelligence. Code, demo and models are available at https://github.com/QwenLM/Qwen-VL.

Abstract: Recent implicit neural representations have shown great results for novel view synthesis. However, existing methods require expensive per-scene optimization from many views hence limiting their application to real-world unbounded urban settings where the objects of interest or backgrounds are observed from very few views. To mitigate this challenge, we introduce a new approach called NeO 360, Neural fields for sparse view synthesis of outdoor scenes. NeO 360 is a generalizable method that reconstructs 360{\deg} scenes from a single or a few posed RGB images. The essence of our approach is in capturing the distribution of complex real-world outdoor 3D scenes and using a hybrid image-conditional triplanar representation that can be queried from any world point. Our representation combines the best of both voxel-based and bird's-eye-view (BEV) representations and is more effective and expressive than each. NeO 360's representation allows us to learn from a large collection of unbounded 3D scenes while offering generalizability to new views and novel scenes from as few as a single image during inference. We demonstrate our approach on the proposed challenging 360{\deg} unbounded dataset, called NeRDS 360, and show that NeO 360 outperforms state-of-the-art generalizable methods for novel view synthesis while also offering editing and composition capabilities. Project page: https://zubair-irshad.github.io/projects/neo360.html

Abstract: Automatic high-quality rendering of anime scenes from complex real-world images is of significant practical value. The challenges of this task lie in the complexity of the scenes, the unique features of anime style, and the lack of high-quality datasets to bridge the domain gap. Despite promising attempts, previous efforts are still incompetent in achieving satisfactory results with consistent semantic preservation, evident stylization, and fine details. In this study, we propose Scenimefy, a novel semi-supervised image-to-image translation framework that addresses these challenges. Our approach guides the learning with structure-consistent pseudo paired data, simplifying the pure unsupervised setting. The pseudo data are derived uniquely from a semantic-constrained StyleGAN leveraging rich model priors like CLIP. We further apply segmentation-guided data selection to obtain high-quality pseudo supervision. A patch-wise contrastive style loss is introduced to improve stylization and fine details. Besides, we contribute a high-resolution anime scene dataset to facilitate future research. Our extensive experiments demonstrate the superiority of our method over state-of-the-art baselines in terms of both perceptual quality and quantitative performance.

Abstract: Existing automatic approaches for 3D virtual character motion synthesis supporting scene interactions do not generalise well to new objects outside training distributions, even when trained on extensive motion capture datasets with diverse objects and annotated interactions. This paper addresses this limitation and shows that robustness and generalisation to novel scene objects in 3D object-aware character synthesis can be achieved by training a motion model with as few as one reference object. We leverage an implicit feature representation trained on object-only datasets, which encodes an SE(3)-equivariant descriptor field around the object. Given an unseen object and a reference pose-object pair, we optimise for the object-aware pose that is closest in the feature space to the reference pose. Finally, we use l-NSM, i.e., our motion generation model that is trained to seamlessly transition from locomotion to object interaction with the proposed bidirectional pose blending scheme. Through comprehensive numerical comparisons to state-of-the-art methods and in a user study, we demonstrate substantial improvements in 3D virtual character motion and interaction quality and robustness to scenarios with unseen objects. Our project page is available at https://vcai.mpi-inf.mpg.de/projects/ROAM/.