In this paper, a novel Diffusion-based 3D Pose estimation (D3DP) method with Joint-wise reProjection-based Multi-hypothesis Aggregation (JPMA) is proposed for probabilistic 3D human pose estimation. On the one hand, D3DP generates multiple possible 3D pose hypotheses for a single 2D observation. It gradually diffuses the ground truth 3D poses to a random distribution, and learns a denoiser conditioned on 2D keypoints to recover the uncontaminated 3D poses. The proposed D3DP is compatible with existing 3D pose estimators and supports users to balance efficiency and accuracy during inference through two customizable parameters. On the other hand, JPMA is proposed to assemble multiple hypotheses generated by D3DP into a single 3D pose for practical use. It reprojects 3D pose hypotheses to the 2D camera plane, selects the best hypothesis joint-by-joint based on the reprojection errors, and combines the selected joints into the final pose. The proposed JPMA conducts aggregation at the joint level and makes use of the 2D prior information, both of which have been overlooked by previous approaches. Extensive experiments on Human3.6M and MPI-INF-3DHP datasets show that our method outperforms the state-of-the-art deterministic and probabilistic approaches by 1.5% and 8.9%, respectively. Code is available at https://github.com/paTRICK-swk/D3DP.
In this paper, we propose a novel framework for Interactive Face Video Coding (IFVC), which allows humans to interact with the intrinsic visual representations instead of the signals. The proposed solution enjoys several distinct advantages, including ultra-compact representation, low delay interaction, and vivid expression and headpose animation. In particular, we propose the Internal Dimension Increase (IDI) based representation, greatly enhancing the fidelity and flexibility in rendering the appearance while maintaining reasonable representation cost. By leveraging strong statistical regularities, the visual signals can be effectively projected into controllable semantics in the three dimensional space (e.g., mouth motion, eye blinking, head rotation and head translation), which are compressed and transmitted. The editable bitstream, which naturally supports the interactivity at the semantic level, can synthesize the face frames via the strong inference ability of the deep generative model. Experimental results have demonstrated the performance superiority and application prospects of our proposed IFVC scheme. In particular, the proposed scheme not only outperforms the state-of-the-art video coding standard Versatile Video Coding (VVC) and the latest generative compression schemes in terms of rate-distortion performance for face videos, but also enables the interactive coding without introducing additional manipulation processes. Furthermore, the proposed framework is expected to shed lights on the future design of the digital human communication in the metaverse.
Skeleton-based human action recognition is a longstanding challenge due to its complex dynamics. Some fine-grain details of the dynamics play a vital role in classification. The existing work largely focuses on designing incremental neural networks with more complicated adjacent matrices to capture the details of joints relationships. However, they still have difficulties distinguishing actions that have broadly similar motion patterns but belong to different categories. Interestingly, we found that the subtle differences in motion patterns can be significantly amplified and become easy for audience to distinct through specified view directions, where this property haven't been fully explored before. Drastically different from previous work, we boost the performance by proposing a conceptually simple yet effective Multi-view strategy that recognizes actions from a collection of dynamic view features. Specifically, we design a novel Skeleton-Anchor Proposal (SAP) module which contains a Multi-head structure to learn a set of views. For feature learning of different views, we introduce a novel Angle Representation to transform the actions under different views and feed the transformations into the baseline model. Our module can work seamlessly with the existing action classification model. Incorporated with baseline models, our SAP module exhibits clear performance gains on many challenging benchmarks. Moreover, comprehensive experiments show that our model consistently beats down the state-of-the-art and remains effective and robust especially when dealing with corrupted data. Related code will be available on https://github.com/ideal-idea/SAP .
We propose a new framework that generalizes the parameters of neural network models to $C^*$-algebra-valued ones. $C^*$-algebra is a generalization of the space of complex numbers. A typical example is the space of continuous functions on a compact space. This generalization enables us to combine multiple models continuously and use tools for functions such as regression and integration. Consequently, we can learn features of data efficiently and adapt the models to problems continuously. We apply our framework to practical problems such as density estimation and few-shot learning and show that our framework enables us to learn features of data even with a limited number of samples. Our new framework highlights the potential possibility of applying the theory of $C^*$-algebra to general neural network models.
Reasoning about causal and temporal event relations in videos is a new destination of Video Question Answering (VideoQA).The major stumbling block to achieve this purpose is the semantic gap between language and video since they are at different levels of abstraction. Existing efforts mainly focus on designing sophisticated architectures while utilizing frame- or object-level visual representations. In this paper, we reconsider the multi-modal alignment problem in VideoQA from feature and sample perspectives to achieve better performance. From the view of feature,we break down the video into trajectories and first leverage trajectory feature in VideoQA to enhance the alignment between two modalities. Moreover, we adopt a heterogeneous graph architecture and design a hierarchical framework to align both trajectory-level and frame-level visual feature with language feature. In addition, we found that VideoQA models are largely dependent on language priors and always neglect visual-language interactions. Thus, two effective yet portable training augmentation strategies are designed to strengthen the cross-modal correspondence ability of our model from the view of sample. Extensive results show that our method outperforms all the state-of-the-art models on the challenging NExT-QA benchmark, which demonstrates the effectiveness of the proposed method.
Supervised federated learning (FL) enables multiple clients to share the trained model without sharing their labeled data. However, potential clients might even be reluctant to label their own data, which could limit the applicability of FL in practice. In this paper, we show the possibility of unsupervised FL whose model is still a classifier for predicting class labels, if the class-prior probabilities are shifted while the class-conditional distributions are shared among the unlabeled data owned by the clients. We propose federation of unsupervised learning (FedUL), where the unlabeled data are transformed into surrogate labeled data for each of the clients, a modified model is trained by supervised FL, and the wanted model is recovered from the modified model. FedUL is a very general solution to unsupervised FL: it is compatible with many supervised FL methods, and the recovery of the wanted model can be theoretically guaranteed as if the data have been labeled. Experiments on benchmark and real-world datasets demonstrate the effectiveness of FedUL. Code is available at https://github.com/lunanbit/FedUL.
Augmented Reality/Virtual Reality (AR/VR) glasses are widely foreseen as the next generation computing platform. AR/VR glasses are a complex "system of systems" which must satisfy stringent form factor, computing-, power- and thermal- requirements. In this paper, we will show that a novel distributed on-sensor compute architecture, coupled with new semiconductor technologies (such as dense 3D-IC interconnects and Spin-Transfer Torque Magneto Random Access Memory, STT-MRAM) and, most importantly, a full hardware-software co-optimization are the solutions to achieve attractive and socially acceptable AR/VR glasses. To this end, we developed a semi-analytical simulation framework to estimate the power consumption of novel AR/VR distributed on-sensor computing architectures. The model allows the optimization of the main technological features of the system modules, as well as the computer-vision algorithm partition strategy across the distributed compute architecture. We show that, in the case of the compute-intensive machine learning based Hand Tracking algorithm, the distributed on-sensor compute architecture can reduce the system power consumption compared to a centralized system, with the additional benefits in terms of latency and privacy.
Skeleton sequences are widely used for action recognition task due to its lightweight and compact characteristics. Recent graph convolutional network (GCN) approaches have achieved great success for skeleton-based action recognition since its grateful modeling ability of non-Euclidean data. GCN is able to utilize the short-range joint dependencies while lack to directly model the distant joints relations that are vital to distinguishing various actions. Thus, many GCN approaches try to employ hierarchical mechanism to aggregate wider-range neighborhood information. We propose a novel self-attention based skeleton-anchor proposal (SAP) module to comprehensively model the internal relations of a human body for motion feature learning. The proposed SAP module aims to explore inherent relationship within human body using a triplet representation via encoding high order angle information rather than the fixed pair-wise bone connection used in the existing hierarchical GCN approaches. A Self-attention based anchor selection method is designed in the proposed SAP module for extracting the root point of encoding angular information. By coupling proposed SAP module with popular spatial-temporal graph neural networks, e.g. MSG3D, it achieves new state-of-the-art accuracy on challenging benchmark datasets. Further ablation study have shown the effectiveness of our proposed SAP module, which is able to obviously improve the performance of many popular skeleton-based action recognition methods.
Recent work has raised concerns on the risk of spurious correlations and unintended biases in statistical machine learning models that threaten model robustness and fairness. In this paper, we propose a simple and intuitive regularization approach to integrate causal knowledge during model training and build a robust and fair model by emphasizing causal features and de-emphasizing spurious features. Specifically, we first manually identify causal and spurious features with principles inspired from the counterfactual framework of causal inference. Then, we propose a regularization approach to penalize causal and spurious features separately. By adjusting the strength of the penalty for each type of feature, we build a predictive model that relies more on causal features and less on non-causal features. We conduct experiments to evaluate model robustness and fairness on three datasets with multiple metrics. Empirical results show that the new models built with causal awareness significantly improve model robustness with respect to counterfactual texts and model fairness with respect to sensitive attributes.
The research of visual signal compression has a long history. Fueled by deep learning, exciting progress has been made recently. Despite achieving better compression performance, existing end-to-end compression algorithms are still designed towards better signal quality in terms of rate-distortion optimization. In this paper, we show that the design and optimization of network architecture could be further improved for compression towards machine vision. We propose an inverted bottleneck structure for end-to-end compression towards machine vision, which specifically accounts for efficient representation of the semantic information. Moreover, we quest the capability of optimization by incorporating the analytics accuracy into the optimization process, and the optimality is further explored with generalized rate-accuracy optimization in an iterative manner. We use object detection as a showcase for end-to-end compression towards machine vision, and extensive experiments show that the proposed scheme achieves significant BD-rate savings in terms of analysis performance. Moreover, the promise of the scheme is also demonstrated with strong generalization capability towards other machine vision tasks, due to the enabling of signal-level reconstruction.