Computer Vision and Pattern Recognition (cs.CV)

Abstract: Combining high-level and low-level visual tasks is a common technique in the field of computer vision. This work integrates the technique of image super resolution to semantic segmentation for document image binarization. It demonstrates that using image super-resolution as a preprocessing step can effectively enhance the results and performance of semantic segmentation.

Abstract: We present a neural network architecture designed to naturally learn a positional embedding and overcome the spectral bias towards lower frequencies faced by conventional implicit neural representation networks. Our proposed architecture, SPDER, is a simple MLP that uses an activation function composed of a sinusoidal multiplied by a sublinear function, called the damping function. The sinusoidal enables the network to automatically learn the positional embedding of an input coordinate while the damping passes on the actual coordinate value by preventing it from being projected down to within a finite range of values. Our results indicate that SPDERs speed up training by 10x and converge to losses 1,500-50,000x lower than that of the state-of-the-art for image representation. SPDER is also state-of-the-art in audio representation. The superior representation capability allows SPDER to also excel on multiple downstream tasks such as image super-resolution and video frame interpolation. We provide intuition as to why SPDER significantly improves fitting compared to that of other INR methods while requiring no hyperparameter tuning or preprocessing.

Abstract: To overcome hardware limitations in commercially available depth sensors which result in low-resolution depth maps, depth map super-resolution (DMSR) is a practical and valuable computer vision task. DMSR requires upscaling a low-resolution (LR) depth map into a high-resolution (HR) space. Joint image filtering for DMSR has been applied using spatially-invariant and spatially-variant convolutional neural network (CNN) approaches. In this project, we propose a novel joint image filtering DMSR algorithm using a Swin transformer architecture. Furthermore, we introduce a Nonlinear Activation Free (NAF) network based on a conventional CNN model used in cutting-edge image restoration applications and compare the performance of the techniques. The proposed algorithms are validated through numerical studies and visual examples demonstrating improvements to state-of-the-art performance while maintaining competitive computation time for noisy depth map super-resolution.

Abstract: In this report, we present our champion solution for Ego4D Natural Language Queries (NLQ) Challenge in CVPR 2023. Essentially, to accurately ground in a video, an effective egocentric feature extractor and a powerful grounding model are required. Motivated by this, we leverage a two-stage pre-training strategy to train egocentric feature extractors and the grounding model on video narrations, and further fine-tune the model on annotated data. In addition, we introduce a novel grounding model GroundNLQ, which employs a multi-modal multi-scale grounding module for effective video and text fusion and various temporal intervals, especially for long videos. On the blind test set, GroundNLQ achieves 25.67 and 18.18 for R1@IoU=0.3 and R1@IoU=0.5, respectively, and surpasses all other teams by a noticeable margin. Our code will be released at\url{https://github.com/houzhijian/GroundNLQ}.

Abstract: Few-shot semantic segmentation (FSS) aims to form class-agnostic models segmenting unseen classes with only a handful of annotations. Previous methods limited to the semantic feature and prototype representation suffer from coarse segmentation granularity and train-set overfitting. In this work, we design Hierarchically Decoupled Matching Network (HDMNet) mining pixel-level support correlation based on the transformer architecture. The self-attention modules are used to assist in establishing hierarchical dense features, as a means to accomplish the cascade matching between query and support features. Moreover, we propose a matching module to reduce train-set overfitting and introduce correlation distillation leveraging semantic correspondence from coarse resolution to boost fine-grained segmentation. Our method performs decently in experiments. We achieve 50.0% mIoU on COCO dataset one-shot setting and 56.0% on five-shot segmentation, respectively. The code will be available on the project website. We hope our work can benefit broader industrial applications where novel classes with limited annotations are required to be decently identified.

Abstract: Transfer learning aims to make the most of existing pre-trained models to achieve better performance on a new task in limited data scenarios. However, it is unclear which models will perform best on which task, and it is prohibitively expensive to try all possible combinations. If transferability estimation offers a computation-efficient approach to evaluate the generalisation ability of models, prior works focused exclusively on classification settings. To overcome this limitation, we extend transferability metrics to object detection. We design a simple method to extract local features corresponding to each object within an image using ROI-Align. We also introduce TLogME, a transferability metric taking into account the coordinates regression task. In our experiments, we compare TLogME to state-of-the-art metrics in the estimation of transfer performance of the Faster-RCNN object detector. We evaluate all metrics on source and target selection tasks, for real and synthetic datasets, and with different backbone architectures. We show that, over different tasks, TLogME using the local extraction method provides a robust correlation with transfer performance and outperforms other transferability metrics on local and global level features.

Abstract: Conventional pedestrian simulators are inevitable tools in the design process of a building, as they enable project engineers to prevent overcrowding situations and plan escape routes for evacuation. However, simulation runtime and the multiple cumbersome steps in generating simulation results are potential bottlenecks during the building design process. Data-driven approaches have demonstrated their capability to outperform conventional methods in speed while delivering similar or even better results across many disciplines. In this work, we present a deep learning-based approach based on a Vision Transformer to predict density heatmaps over time and total evacuation time from a given floorplan. Specifically, due to limited availability of public datasets, we implement a parametric data generation pipeline including a conventional simulator. This enables us to build a large synthetic dataset that we use to train our architecture. Furthermore, we seamlessly integrate our model into a BIM-authoring tool to generate simulation results instantly and automatically.

Abstract: Graph convolutional networks have been widely used in skeleton-based action recognition. However, existing approaches are limited in fine-grained action recognition due to the similarity of inter-class data. Moreover, the noisy data from pose extraction increases the challenge of fine-grained recognition. In this work, we propose a flexible attention block called Channel-Variable Spatial-Temporal Attention (CVSTA) to enhance the discriminative power of spatial-temporal joints and obtain a more compact intra-class feature distribution. Based on CVSTA, we construct a Multi-Dimensional Refinement Graph Convolutional Network (MDR-GCN), which can improve the discrimination among channel-, joint- and frame-level features for fine-grained actions. Furthermore, we propose a Robust Decouple Loss (RDL), which significantly boosts the effect of the CVSTA and reduces the impact of noise. The proposed method combining MDR-GCN with RDL outperforms the known state-of-the-art skeleton-based approaches on fine-grained datasets, FineGym99 and FSD-10, and also on the coarse dataset NTU-RGB+D X-view version.

Abstract: This paper proposes Shoggoth, an efficient edge-cloud collaborative architecture, for boosting inference performance on real-time video of changing scenes. Shoggoth uses online knowledge distillation to improve the accuracy of models suffering from data drift and offloads the labeling process to the cloud, alleviating constrained resources of edge devices. At the edge, we design adaptive training using small batches to adapt models under limited computing power, and adaptive sampling of training frames for robustness and reducing bandwidth. The evaluations on the realistic dataset show 15%-20% model accuracy improvement compared to the edge-only strategy and fewer network costs than the cloud-only strategy.

Abstract: Reliable LiDAR panoptic segmentation (LPS), including both semantic and instance segmentation, is vital for many robotic applications, such as autonomous driving. This work proposes a new LPS framework named PANet to eliminate the dependency on the offset branch and improve the performance on large objects, which are always over-segmented by clustering algorithms. Firstly, we propose a non-learning Sparse Instance Proposal (SIP) module with the ``sampling-shifting-grouping" scheme to directly group thing points into instances from the raw point cloud efficiently. More specifically, balanced point sampling is introduced to generate sparse seed points with more uniform point distribution over the distance range. And a shift module, termed bubble shifting, is proposed to shrink the seed points to the clustered centers. Then we utilize the connected component label algorithm to generate instance proposals. Furthermore, an instance aggregation module is devised to integrate potentially fragmented instances, improving the performance of the SIP module on large objects. Extensive experiments show that PANet achieves state-of-the-art performance among published works on the SemanticKITII validation and nuScenes validation for the panoptic segmentation task.

Abstract: Semantic scene completion (SSC) jointly predicts the semantics and geometry of the entire 3D scene, which plays an essential role in 3D scene understanding for autonomous driving systems. SSC has achieved rapid progress with the help of semantic context in segmentation. However, how to effectively exploit the relationships between the semantic context in semantic segmentation and geometric structure in scene completion remains under exploration. In this paper, we propose to solve outdoor SSC from the perspective of representation separation and BEV fusion. Specifically, we present the network, named SSC-RS, which uses separate branches with deep supervision to explicitly disentangle the learning procedure of the semantic and geometric representations. And a BEV fusion network equipped with the proposed Adaptive Representation Fusion (ARF) module is presented to aggregate the multi-scale features effectively and efficiently. Due to the low computational burden and powerful representation ability, our model has good generality while running in real-time. Extensive experiments on SemanticKITTI demonstrate our SSC-RS achieves state-of-the-art performance.

Abstract: Space-time memory (STM) network methods have been dominant in semi-supervised video object segmentation (SVOS) due to their remarkable performance. In this work, we identify three key aspects where we can improve such methods; i) supervisory signal, ii) pretraining and iii) spatial awareness. We then propose TrickVOS; a generic, method-agnostic bag of tricks addressing each aspect with i) a structure-aware hybrid loss, ii) a simple decoder pretraining regime and iii) a cheap tracker that imposes spatial constraints in model predictions. Finally, we propose a lightweight network and show that when trained with TrickVOS, it achieves competitive results to state-of-the-art methods on DAVIS and YouTube benchmarks, while being one of the first STM-based SVOS methods that can run in real-time on a mobile device.

Abstract: Neural architecture search (NAS) proves to be among the effective approaches for many tasks by generating an application-adaptive neural architecture, which is still challenged by high computational cost and memory consumption. At the same time, 1-bit convolutional neural networks (CNNs) with binary weights and activations show their potential for resource-limited embedded devices. One natural approach is to use 1-bit CNNs to reduce the computation and memory cost of NAS by taking advantage of the strengths of each in a unified framework, while searching the 1-bit CNNs is more challenging due to the more complicated processes involved. In this paper, we introduce Discrepant Child-Parent Neural Architecture Search (DCP-NAS) to efficiently search 1-bit CNNs, based on a new framework of searching the 1-bit model (Child) under the supervision of a real-valued model (Parent). Particularly, we first utilize a Parent model to calculate a tangent direction, based on which the tangent propagation method is introduced to search the optimized 1-bit Child. We further observe a coupling relationship between the weights and architecture parameters existing in such differentiable frameworks. To address the issue, we propose a decoupled optimization method to search an optimized architecture. Extensive experiments demonstrate that our DCP-NAS achieves much better results than prior arts on both CIFAR-10 and ImageNet datasets. In particular, the backbones achieved by our DCP-NAS achieve strong generalization performance on person re-identification and object detection.

Abstract: Lane graph estimation is a long-standing problem in the context of autonomous driving. Previous works aimed at solving this problem by relying on large-scale, hand-annotated lane graphs, introducing a data bottleneck for training models to solve this task. To overcome this limitation, we propose to use the motion patterns of traffic participants as lane graph annotations. In our AutoGraph approach, we employ a pre-trained object tracker to collect the tracklets of traffic participants such as vehicles and trucks. Based on the location of these tracklets, we predict the successor lane graph from an initial position using overhead RGB images only, not requiring any human supervision. In a subsequent stage, we show how the individual successor predictions can be aggregated into a consistent lane graph. We demonstrate the efficacy of our approach on the UrbanLaneGraph dataset and perform extensive quantitative and qualitative evaluations, indicating that AutoGraph is on par with models trained on hand-annotated graph data. Model and dataset will be made available at redacted-for-review.

Abstract: Change detection and irregular object extraction in 3D point clouds is a challenging task that is of high importance not only for autonomous navigation but also for updating existing digital twin models of various industrial environments. This article proposes an innovative approach for change detection in 3D point clouds using deep learned place recognition descriptors and irregular object extraction based on voxel-to-point comparison. The proposed method first aligns the bi-temporal point clouds using a map-merging algorithm in order to establish a common coordinate frame. Then, it utilizes deep learning techniques to extract robust and discriminative features from the 3D point cloud scans, which are used to detect changes between consecutive point cloud frames and therefore find the changed areas. Finally, the altered areas are sampled and compared between the two time instances to extract any obstructions that caused the area to change. The proposed method was successfully evaluated in real-world field experiments, where it was able to detect different types of changes in 3D point clouds, such as object or muck-pile addition and displacement, showcasing the effectiveness of the approach. The results of this study demonstrate important implications for various applications, including safety and security monitoring in construction sites, mapping and exploration and suggests potential future research directions in this field.

Abstract: Efficiently generating a free-style 3D portrait with high quality and consistency is a promising yet challenging task. The portrait styles generated by most existing methods are usually restricted by their 3D generators, which are learned in specific facial datasets, such as FFHQ. To get a free-style 3D portrait, one can build a large-scale multi-style database to retrain the 3D generator, or use a off-the-shelf tool to do the style translation. However, the former is time-consuming due to data collection and training process, the latter may destroy the multi-view consistency. To tackle this problem, we propose a fast 3D portrait synthesis framework in this paper, which enable one to use text prompts to specify styles. Specifically, for a given portrait style, we first leverage two generative priors, a 3D-aware GAN generator (EG3D) and a text-guided image editor (Ip2p), to quickly construct a few-shot training set, where the inference process of Ip2p is optimized to make editing more stable. Then we replace original triplane generator of EG3D with a Image-to-Triplane (I2T) module for two purposes: 1) getting rid of the style constraints of pre-trained EG3D by fine-tuning I2T on the few-shot dataset; 2) improving training efficiency by fixing all parts of EG3D except I2T. Furthermore, we construct a multi-style and multi-identity 3D portrait database to demonstrate the scalability and generalization of our method. Experimental results show that our method is capable of synthesizing high-quality 3D portraits with specified styles in a few minutes, outperforming the state-of-the-art.

Abstract: No-service rail surface defect (NRSD) segmentation is an essential way for perceiving the quality of no-service rails. However, due to the complex and diverse outlines and low-contrast textures of no-service rails, existing natural image segmentation methods cannot achieve promising performance in NRSD images, especially in some unique and challenging NRSD scenes. To this end, in this paper, we propose a novel segmentation network for NRSDs based on Normalized Attention and Dual-scale Interaction, named NaDiNet. Specifically, NaDiNet follows the enhancement-interaction paradigm. The Normalized Channel-wise Self-Attention Module (NAM) and the Dual-scale Interaction Block (DIB) are two key components of NaDiNet. NAM is a specific extension of the channel-wise self-attention mechanism (CAM) to enhance features extracted from low-contrast NRSD images. The softmax layer in CAM will produce very small correlation coefficients which are not conducive to low-contrast feature enhancement. Instead, in NAM, we directly calculate the normalized correlation coefficient between channels to enlarge the feature differentiation. DIB is specifically designed for the feature interaction of the enhanced features. It has two interaction branches with dual scales, one for fine-grained clues and the other for coarse-grained clues. With both branches working together, DIB can perceive defect regions of different granularities. With these modules working together, our NaDiNet can generate accurate segmentation map. Extensive experiments on the public NRSD-MN dataset with man-made and natural NRSDs demonstrate that our proposed NaDiNet with various backbones (i.e., VGG, ResNet, and DenseNet) consistently outperforms 10 state-of-the-art methods. The code and results of our method are available at https://github.com/monxxcn/NaDiNet.

Abstract: In this report, we present the technical details of our submission to the EPIC-Kitchens-100 Multi-Instance Retrieval Challenge 2023. To participate in the challenge, we ensembled two models trained with two different loss functions on 25% of the training data. Our submission, visible on the public leaderboard, obtains an average score of 56.81% nDCG and 42.63% mAP.

Abstract: Recent works have demonstrated that deep neural networks (DNNs) are highly vulnerable to adversarial attacks. To defend against adversarial attacks, many defense strategies have been proposed, among which adversarial training has been demonstrated to be the most effective strategy. However, it has been known that adversarial training sometimes hurts natural accuracy. Then, many works focus on optimizing model parameters to handle the problem. Different from the previous approaches, in this paper, we propose a new approach to improve the adversarial robustness by using an external signal rather than model parameters. In the proposed method, a well-optimized universal external signal called a booster signal is injected into the outside of the image which does not overlap with the original content. Then, it boosts both adversarial robustness and natural accuracy. The booster signal is optimized in parallel to model parameters step by step collaboratively. Experimental results show that the booster signal can improve both the natural and robust accuracies over the recent state-of-the-art adversarial training methods. Also, optimizing the booster signal is general and flexible enough to be adopted on any existing adversarial training methods.

Abstract: Recently, it has been widely known that deep neural networks are highly vulnerable and easily broken by adversarial attacks. To mitigate the adversarial vulnerability, many defense algorithms have been proposed. Recently, to improve adversarial robustness, many works try to enhance feature representation by imposing more direct supervision on the discriminative feature. However, existing approaches lack an understanding of learning adversarially robust feature representation. In this paper, we propose a novel training framework called Robust Proxy Learning. In the proposed method, the model explicitly learns robust feature representations with robust proxies. To this end, firstly, we demonstrate that we can generate class-representative robust features by adding class-wise robust perturbations. Then, we use the class representative features as robust proxies. With the class-wise robust features, the model explicitly learns adversarially robust features through the proposed robust proxy learning framework. Through extensive experiments, we verify that we can manually generate robust features, and our proposed learning framework could increase the robustness of the DNNs.

Abstract: The accurate detection of suspicious regions in medical images is an error-prone and time-consuming process required by many routinely performed diagnostic procedures. To support clinicians during this difficult task, several automated solutions were proposed relying on complex methods with many hyperparameters. In this study, we investigate the feasibility of DEtection TRansformer (DETR) models for volumetric medical object detection. In contrast to previous works, these models directly predict a set of objects without relying on the design of anchors or manual heuristics such as non-maximum-suppression to detect objects. We show by conducting extensive experiments with three models, namely DETR, Conditional DETR, and DINO DETR on four data sets (CADA, RibFrac, KiTS19, and LIDC) that these set prediction models can perform on par with or even better than currently existing methods. DINO DETR, the best-performing model in our experiments demonstrates this by outperforming a strong anchor-based one-stage detector, Retina U-Net, on three out of four data sets.

Abstract: This paper makes the first attempt to tackle the challenging task of recovering arbitrary frame rate latent global shutter (GS) frames from two consecutive rolling shutter (RS) frames, guided by the novel event camera data. Although events possess high temporal resolution, beneficial for video frame interpolation (VFI), a hurdle in tackling this task is the lack of paired GS frames. Another challenge is that RS frames are susceptible to distortion when capturing moving objects. To this end, we propose a novel self-supervised framework that leverages events to guide RS frame correction and VFI in a unified framework. Our key idea is to estimate the displacement field (DF) non-linear dense 3D spatiotemporal information of all pixels during the exposure time, allowing for the reciprocal reconstruction between RS and GS frames as well as arbitrary frame rate VFI. Specifically, the displacement field estimation (DFE) module is proposed to estimate the spatiotemporal motion from events to correct the RS distortion and interpolate the GS frames in one step. We then combine the input RS frames and DF to learn a mapping for RS-to-GS frame interpolation. However, as the mapping is highly under-constrained, we couple it with an inverse mapping (i.e., GS-to-RS) and RS frame warping (i.e., RS-to-RS) for self-supervision. As there is a lack of labeled datasets for evaluation, we generate two synthetic datasets and collect a real-world dataset to train and test our method. Experimental results show that our method yields comparable or better performance with prior supervised methods.

Abstract: Abdominal multi-organ segmentation from CT and MRI is an essential prerequisite for surgical planning and computer-aided navigation systems. Three-dimensional numeric representations of abdominal shapes are further important for quantitative and statistical analyses thereof. Existing methods in the field, however, are unable to extract highly accurate 3D representations that are smooth, topologically correct, and match points on a template. In this work, we present UNetFlow, a novel diffeomorphic shape deformation approach for abdominal organs. UNetFlow combines the advantages of voxel-based and mesh-based approaches for 3D shape extraction. Our results demonstrate high accuracy with respect to manually annotated CT data and better topological correctness compared to previous methods. In addition, we show the generalization of UNetFlow to MRI.

Abstract: While semantic segmentation has seen tremendous improvements in the past, there is still significant labeling efforts necessary and the problem of limited generalization to classes that have not been present during training. To address this problem, zero-shot semantic segmentation makes use of large self-supervised vision-language models, allowing zero-shot transfer to unseen classes. In this work, we build a benchmark for Multi-domain Evaluation of Semantic Segmentation (MESS), which allows a holistic analysis of performance across a wide range of domain-specific datasets such as medicine, engineering, earth monitoring, biology, and agriculture. To do this, we reviewed 120 datasets, developed a taxonomy, and classified the datasets according to the developed taxonomy. We select a representative subset consisting of 22 datasets and propose it as the MESS benchmark. We evaluate eight recently published models on the proposed MESS benchmark and analyze characteristics for the performance of zero-shot transfer models. The toolkit is available at https://github.com/blumenstiel/MESS.

Abstract: Contrast-Enhanced Ultra-Sound (CEUS) has become a viable method for non-invasive, dynamic visualization in medical diagnostics, yet Ultrasound Localization Microscopy (ULM) has enabled a revolutionary breakthrough by offering ten times higher resolution. To date, Delay-And-Sum (DAS) beamformers are used to render ULM frames, ultimately determining the image resolution capability. To take full advantage of ULM, this study questions whether beamforming is the most effective processing step for ULM, suggesting an alternative approach that relies solely on Time-Difference-of-Arrival (TDoA) information. To this end, a novel geometric framework for micro bubble localization via ellipse intersections is proposed to overcome existing beamforming limitations. We present a benchmark comparison based on a public dataset for which our geometric ULM outperforms existing baseline methods in terms of accuracy and reliability while only utilizing a portion of the available transducer data.

Abstract: Learned image compression (LIC) methods have experienced significant progress during recent years. However, these methods are primarily dedicated to optimizing the rate-distortion (R-D) performance at medium and high bitrates (> 0.1 bits per pixel (bpp)), while research on extremely low bitrates is limited. Besides, existing methods fail to explicitly explore the image structure and texture components crucial for image compression, treating them equally alongside uninformative components in networks. This can cause severe perceptual quality degradation, especially under low-bitrate scenarios. In this work, inspired by the success of pre-trained masked autoencoders (MAE) in many downstream tasks, we propose to rethink its mask sampling strategy from structure and texture perspectives for high redundancy reduction and discriminative feature representation, further unleashing the potential of LIC methods. Therefore, we present a dual-adaptive masking approach (DA-Mask) that samples visible patches based on the structure and texture distributions of original images. We combine DA-Mask and pre-trained MAE in masked image modeling (MIM) as an initial compressor that abstracts informative semantic context and texture representations. Such a pipeline can well cooperate with LIC networks to achieve further secondary compression while preserving promising reconstruction quality. Consequently, we propose a simple yet effective masked compression model (MCM), the first framework that unifies MIM and LIC end-to-end for extremely low-bitrate image compression. Extensive experiments have demonstrated that our approach outperforms recent state-of-the-art methods in R-D performance, visual quality, and downstream applications, at very low bitrates. Our code is available at https://github.com/lianqi1008/MCM.git.

Abstract: In recent years, deep learning models have revolutionized medical image interpretation, offering substantial improvements in diagnostic accuracy. However, these models often struggle with challenging images where critical features are partially or fully occluded, which is a common scenario in clinical practice. In this paper, we propose a novel curriculum learning-based approach to train deep learning models to handle occluded medical images effectively. Our method progressively introduces occlusion, starting from clear, unobstructed images and gradually moving to images with increasing occlusion levels. This ordered learning process, akin to human learning, allows the model to first grasp simple, discernable patterns and subsequently build upon this knowledge to understand more complicated, occluded scenarios. Furthermore, we present three novel occlusion synthesis methods, namely Wasserstein Curriculum Learning (WCL), Information Adaptive Learning (IAL), and Geodesic Curriculum Learning (GCL). Our extensive experiments on diverse medical image datasets demonstrate substantial improvements in model robustness and diagnostic accuracy over conventional training methodologies.

Abstract: Features of pericoronary adipose tissue (PCAT) assessed from coronary computed tomography angiography (CCTA) are associated with inflammation and cardiovascular risk. As PCAT is vascularly connected with coronary vasculature, the presence of iodine is a potential confounding factor on PCAT HU and textures that has not been adequately investigated. Use dynamic cardiac CT perfusion (CCTP) to inform contrast determinants of PCAT assessment. From CCTP, we analyzed HU dynamics of territory-specific PCAT, myocardium, and other adipose depots in patients with coronary artery disease. HU, blood flow, and radiomics were assessed over time. Changes from peak aorta time, Pa, chosen to model the time of CCTA, were obtained. HU in PCAT increased more than in other adipose depots. The estimated blood flow in PCAT was ~23% of that in the contiguous myocardium. Comparing PCAT distal and proximal to a significant stenosis, we found less enhancement and longer time-to-peak distally. Two-second offsets [before, after] Pa resulted in [ 4-HU, 3-HU] differences in PCAT. Due to changes in HU, the apparent PCAT volume reduced ~15% from the first scan (P1) to Pa using a conventional fat window. Comparing radiomic features over time, 78% of features changed >10% relative to P1. CCTP elucidates blood flow in PCAT and enables analysis of PCAT features over time. PCAT assessments (HU, apparent volume, and radiomics) are sensitive to acquisition timing and the presence of obstructive stenosis, which may confound the interpretation of PCAT in CCTA images. Data normalization may be in order.

Abstract: Despite the great success of deep learning in stereo matching, recovering accurate and clearly-contoured disparity map is still challenging. Currently, L1 loss and cross-entropy loss are the two most widely used loss functions for training the stereo matching networks. Comparing with the former, the latter can usually achieve better results thanks to its direct constraint to the the cost volume. However, how to generate reasonable ground-truth distribution for this loss function remains largely under exploited. Existing works assume uni-modal distributions around the ground-truth for all of the pixels, which ignores the fact that the edge pixels may have multi-modal distributions. In this paper, we first experimentally exhibit the importance of correct edge supervision to the overall disparity accuracy. Then a novel adaptive multi-modal cross-entropy loss which encourages the network to generate different distribution patterns for edge and non-edge pixels is proposed. We further optimize the disparity estimator in the inference stage to alleviate the bleeding and misalignment artifacts at the edge. Our method is generic and can help classic stereo matching models regain competitive performance. GANet trained by our loss ranks 1st on the KITTI 2015 and 2012 benchmarks and outperforms state-of-the-art methods by a large margin. Meanwhile, our method also exhibits superior cross-domain generalization ability and outperforms existing generalization-specialized methods on four popular real-world datasets.

Abstract: The recent work CLIPA presents an inverse scaling law for CLIP training -- whereby the larger the image/text encoders used, the shorter the sequence length of image/text tokens that can be applied in training. This finding enables us to train high-performance CLIP models with significantly reduced computations. Building upon this work, we hereby present CLIPA-v2 with two key contributions. Technically, we find this inverse scaling law is also applicable in the finetuning stage, enabling further reduction in computational needs. Empirically, we explore CLIPA at scale, extending the experiments up to the H/14 model with ~13B image-text pairs seen during training. Our results are exciting -- by only allocating a budget of \$10,000, our CLIP model achieves an impressive zero-shot ImageNet accuracy of 81.1%, surpassing the prior best CLIP model (from OpenCLIP, 80.1%) by 1.0% and meanwhile reducing the computational cost by ~39X. Moreover, with an additional investment of $4,000, we can further elevate the zero-shot ImageNet accuracy to 81.8%. Our code and models are available at https://github.com/UCSC-VLAA/CLIPA.

Abstract: Intrinsic image decomposition and inverse rendering are long-standing problems in computer vision. To evaluate albedo recovery, most algorithms report their quantitative performance with a mean Weighted Human Disagreement Rate (WHDR) metric on the IIW dataset. However, WHDR focuses only on relative albedo values and often fails to capture overall quality of the albedo. In order to comprehensively evaluate albedo, we collect a new dataset, Measured Albedo in the Wild (MAW), and propose three new metrics that complement WHDR: intensity, chromaticity and texture metrics. We show that existing algorithms often improve WHDR metric but perform poorly on other metrics. We then finetune different algorithms on our MAW dataset to significantly improve the quality of the reconstructed albedo both quantitatively and qualitatively. Since the proposed intensity, chromaticity, and texture metrics and the WHDR are all complementary we further introduce a relative performance measure that captures average performance. By analysing existing algorithms we show that there is significant room for improvement. Our dataset and evaluation metrics will enable researchers to develop algorithms that improve albedo reconstruction. Code and Data available at: https://measuredalbedo.github.io/

Abstract: Camera pose estimation is a long-standing computer vision problem that to date often relies on classical methods, such as handcrafted keypoint matching, RANSAC and bundle adjustment. In this paper, we propose to formulate the Structure from Motion (SfM) problem inside a probabilistic diffusion framework, modelling the conditional distribution of camera poses given input images. This novel view of an old problem has several advantages. (i) The nature of the diffusion framework mirrors the iterative procedure of bundle adjustment. (ii) The formulation allows a seamless integration of geometric constraints from epipolar geometry. (iii) It excels in typically difficult scenarios such as sparse views with wide baselines. (iv) The method can predict intrinsics and extrinsics for an arbitrary amount of images. We demonstrate that our method PoseDiffusion significantly improves over the classic SfM pipelines and the learned approaches on two real-world datasets. Finally, it is observed that our method can generalize across datasets without further training. Project page: https://posediffusion.github.io/

Abstract: General physical scene understanding requires more than simply localizing and recognizing objects -- it requires knowledge that objects can have different latent properties (e.g., mass or elasticity), and that those properties affect the outcome of physical events. While there has been great progress in physical and video prediction models in recent years, benchmarks to test their performance typically do not require an understanding that objects have individual physical properties, or at best test only those properties that are directly observable (e.g., size or color). This work proposes a novel dataset and benchmark, termed Physion++, that rigorously evaluates visual physical prediction in artificial systems under circumstances where those predictions rely on accurate estimates of the latent physical properties of objects in the scene. Specifically, we test scenarios where accurate prediction relies on estimates of properties such as mass, friction, elasticity, and deformability, and where the values of those properties can only be inferred by observing how objects move and interact with other objects or fluids. We evaluate the performance of a number of state-of-the-art prediction models that span a variety of levels of learning vs. built-in knowledge, and compare that performance to a set of human predictions. We find that models that have been trained using standard regimes and datasets do not spontaneously learn to make inferences about latent properties, but also that models that encode objectness and physical states tend to make better predictions. However, there is still a huge gap between all models and human performance, and all models' predictions correlate poorly with those made by humans, suggesting that no state-of-the-art model is learning to make physical predictions in a human-like way. Project page: https://dingmyu.github.io/physion_v2/

Abstract: We propose a new structure-from-motion framework to recover accurate camera poses and point clouds from unordered images. Traditional SfM systems typically rely on the successful detection of repeatable keypoints across multiple views as the first step, which is difficult for texture-poor scenes, and poor keypoint detection may break down the whole SfM system. We propose a new detector-free SfM framework to draw benefits from the recent success of detector-free matchers to avoid the early determination of keypoints, while solving the multi-view inconsistency issue of detector-free matchers. Specifically, our framework first reconstructs a coarse SfM model from quantized detector-free matches. Then, it refines the model by a novel iterative refinement pipeline, which iterates between an attention-based multi-view matching module to refine feature tracks and a geometry refinement module to improve the reconstruction accuracy. Experiments demonstrate that the proposed framework outperforms existing detector-based SfM systems on common benchmark datasets. We also collect a texture-poor SfM dataset to demonstrate the capability of our framework to reconstruct texture-poor scenes. Based on this framework, we take $\textit{first place}$ in Image Matching Challenge 2023.

Abstract: 3D Semantic Scene Completion (SSC) has emerged as a nascent and pivotal task for autonomous driving, as it involves predicting per-voxel occupancy within a 3D scene from partial LiDAR or image inputs. Existing methods primarily focus on the voxel-wise feature aggregation, while neglecting the instance-centric semantics and broader context. In this paper, we present a novel paradigm termed Symphonies (Scene-from-Insts) for SSC, which completes the scene volume from a sparse set of instance queries derived from the input with context awareness. By incorporating the queries as the instance feature representations within the scene, Symphonies dynamically encodes the instance-centric semantics to interact with the image and volume features while avoiding the dense voxel-wise modeling. Simultaneously, it orchestrates a more comprehensive understanding of the scenario by capturing context throughout the entire scene, contributing to alleviating the geometric ambiguity derived from occlusion and perspective errors. Symphonies achieves a state-of-the-art result of 13.02 mIoU on the challenging SemanticKITTI dataset, outperforming existing methods and showcasing the promising advancements of the paradigm. The code is available at \url{https://github.com/hustvl/Symphonies}.