Estimating the articulated 3D hand-object pose from a single RGB image is a highly ambiguous and challenging problem requiring large-scale datasets that contain diverse hand poses, object poses, and camera viewpoints. Most real-world datasets lack this diversity. In contrast, synthetic datasets can easily ensure vast diversity, but learning from them is inefficient and suffers from heavy training consumption. To address the above issues, we propose ArtiBoost, a lightweight online data enrichment method that boosts articulated hand-object pose estimation from the data perspective. ArtiBoost is employed along with a real-world source dataset. During training, ArtiBoost alternatively performs data exploration and synthesis. ArtiBoost can cover various hand-object poses and camera viewpoints based on a Compositional hand-object Configuration and Viewpoint space (CCV-space) and can adaptively enrich the current hard-discernable samples by a mining strategy. We apply ArtiBoost on a simple learning baseline network and demonstrate the performance boost on several hand-object benchmarks. As an illustrative example, with ArtiBoost, even a simple baseline network can outperform the previous start-of-the-art based on Transformer on the HO3D dataset. Our code is available at https://github.com/MVIG-SJTU/ArtiBoost.
Heatmap-based methods dominate in the field of human pose estimation by modelling the output distribution through likelihood heatmaps. In contrast, regression-based methods are more efficient but suffer from inferior performance. In this work, we explore maximum likelihood estimation (MLE) to develop an efficient and effective regression-based methods. From the perspective of MLE, adopting different regression losses is making different assumptions about the output density function. A density function closer to the true distribution leads to a better regression performance. In light of this, we propose a novel regression paradigm with Residual Log-likelihood Estimation (RLE) to capture the underlying output distribution. Concretely, RLE learns the change of the distribution instead of the unreferenced underlying distribution to facilitate the training process. With the proposed reparameterization design, our method is compatible with off-the-shelf flow models. The proposed method is effective, efficient and flexible. We show its potential in various human pose estimation tasks with comprehensive experiments. Compared to the conventional regression paradigm, regression with RLE bring 12.4 mAP improvement on MSCOCO without any test-time overhead. Moreover, for the first time, especially on multi-person pose estimation, our regression method is superior to the heatmap-based methods. Our code is available at https://github.com/Jeff-sjtu/res-loglikelihood-regression
Direct contour regression for instance segmentation is a challenging task. Previous works usually achieve it by learning to progressively refine the contour prediction or adopting a shape representation with limited expressiveness. In this work, we argue that the difficulty in regressing the contour points in one pass is mainly due to the ambiguity when discretizing a smooth contour into a polygon. To address the ambiguity, we propose a novel differentiable rendering-based approach named \textbf{ContourRender}. During training, it first predicts a contour generated by an invertible shape signature, and then optimizes the contour with the more stable silhouette by converting it to a contour mesh and rendering the mesh to a 2D map. This method significantly improves the quality of contour without iterations or cascaded refinements. Moreover, as optimization is not needed during inference, the inference speed will not be influenced. Experiments show the proposed ContourRender outperforms all the contour-based instance segmentation approaches on COCO, while stays competitive with the iteration-based state-of-the-art on Cityscapes. In addition, we specifically select a subset from COCO val2017 named COCO ContourHard-val to further demonstrate the contour quality improvements. Codes, models, and dataset split will be released.
Human life is populated with articulated objects. Current Category-level Articulation Pose Estimation (CAPE) methods are studied under the single-instance setting with a fixed kinematic structure for each category. Considering these limitations, we reform this problem setting for real-world environments and suggest a CAPE-Real (CAPER) task setting. This setting allows varied kinematic structures within a semantic category, and multiple instances to co-exist in an observation of real world. To support this task, we build an articulated model repository ReArt-48 and present an efficient dataset generation pipeline, which contains Fast Articulated Object Modeling (FAOM) and Semi-Authentic MixEd Reality Technique (SAMERT). Accompanying the pipeline, we build a large-scale mixed reality dataset ReArtMix and a real world dataset ReArtVal. We also propose an effective framework ReArtNOCS that exploits RGB-D input to estimate part-level pose for multiple instances in a single forward pass. Extensive experiments demonstrate that the proposed ReArtNOCS can achieve good performance on both CAPER and CAPE settings. We believe it could serve as a strong baseline for future research on the CAPER task.
Object handover is a common human collaboration behavior that attracts attention from researchers in Robotics and Cognitive Science. Though visual perception plays an important role in the object handover task, the whole handover process has been specifically explored. In this work, we propose a novel rich-annotated dataset, H2O, for visual analysis of human-human object handovers. The H2O, which contains 18K video clips involving 15 people who hand over 30 objects to each other, is a multi-purpose benchmark. It can support several vision-based tasks, from which, we specifically provide a baseline method, RGPNet, for a less-explored task named Receiver Grasp Prediction. Extensive experiments show that the RGPNet can produce plausible grasps based on the giver's hand-object states in the pre-handover phase. Besides, we also report the hand and object pose errors with existing baselines and show that the dataset can serve as the video demonstrations for robot imitation learning on the handover task. Dataset, model and code will be made public.
Video action recognition (VAR) is a primary task of video understanding, and untrimmed videos are more common in real-life scenes. Untrimmed videos have redundant and diverse clips containing contextual information, so sampling dense clips is essential. Recently, some works attempt to train a generic model to select the N most representative clips. However, it is difficult to model the complex relations from intra-class clips and inter-class videos within a single model and fixed selected number, and the entanglement of multiple relations is also hard to explain. Thus, instead of "only look once", we argue "divide and conquer" strategy will be more suitable in untrimmed VAR. Inspired by the speed reading mechanism, we propose a simple yet effective clip-level solution based on skim-scan techniques. Specifically, the proposed Skim-Scan framework first skims the entire video and drops those uninformative and misleading clips. For the remaining clips, it scans clips with diverse features gradually to drop redundant clips but cover essential content. The above strategies can adaptively select the necessary clips according to the difficulty of the different videos. To trade off the computational complexity and performance, we observe the similar statistical expression between lightweight and heavy networks, thus it supports us to explore the combination of them. Comprehensive experiments are performed on ActivityNet and mini-FCVID datasets, and results demonstrate that our solution surpasses the state-of-the-art performance in terms of both accuracy and efficiency.
Suction is an important solution for the longstanding robotic grasping problem. Compared with other kinds of grasping, suction grasping is easier to represent and often more reliable in practice. Though preferred in many scenarios, it is not fully investigated and lacks sufficient training data and evaluation benchmarks. To address that, firstly, we propose a new physical model to analytically evaluate seal formation and wrench resistance of a suction grasping, which are two key aspects of grasp success. Secondly, a two-step methodology is adopted to generate annotations on a large-scale dataset collected in real-world cluttered scenarios. Thirdly, a standard online evaluation system is proposed to evaluate suction poses in continuous operation space, which can benchmark different algorithms fairly without the need of exhaustive labeling. Real-robot experiments are conducted to show that our annotations align well with real world. Meanwhile, we propose a method to predict numerous suction poses from an RGB-D image of a cluttered scene and demonstrate our superiority against several previous methods. Result analyses are further provided to help readers better understand the challenges in this area. Data and source code are publicly available at www.graspnet.net.
Multimodal prediction results are essential for trajectory prediction task as there is no single correct answer for the future. Previous frameworks can be divided into three categories: regression, generation and classification frameworks. However, these frameworks have weaknesses in different aspects so that they cannot model the multimodal prediction task comprehensively. In this paper, we present a novel insight along with a brand-new prediction framework by formulating multimodal prediction into three steps: modality clustering, classification and synthesis, and address the shortcomings of earlier frameworks. Exhaustive experiments on popular benchmarks have demonstrated that our proposed method surpasses state-of-the-art works even without introducing social and map information. Specifically, we achieve 19.2% and 20.8% improvement on ADE and FDE respectively on ETH/UCY dataset. Our code will be made publicly availabe.
Convolutional video models have an order of magnitude larger computational complexity than their counterpart image-level models. Constrained by computational resources, there is no model or training method that can train long video sequences end-to-end. Currently, the main-stream method is to split a raw video into clips, leading to incomplete fragmentary temporal information flow. Inspired by natural language processing techniques dealing with long sentences, we propose to treat videos as serial fragments satisfying Markov property, and train it as a whole by progressively propagating information through the temporal dimension in multiple steps. This progressive training (PGT) method is able to train long videos end-to-end with limited resources and ensures the effective transmission of information. As a general and robust training method, we empirically demonstrate that it yields significant performance improvements on different models and datasets. As an illustrative example, the proposed method improves SlowOnly network by 3.7 mAP on Charades and 1.9 top-1 accuracy on Kinetics with negligible parameter and computation overhead. Code is available at https://github.com/BoPang1996/PGT.