We present a conceptually simple self-supervised method for saliency detection. Our method generates and uses pseudo-ground truth labels for training. The generated pseudo-GT labels don’t require any kind of human annotations (e.g., pixel-wise labels or weak labels like scribbles). Recent works show that features extracted from classification tasks provide important saliency cues like structure and semantic information of salient objects in the image. Our method, called 3SD, exploits this idea by adding a branch for a self-supervised classification task in parallel with salient object detection, to obtain class activation maps (CAM maps). These CAM maps along with the edges of the input image are used to generate the pseudo-GT saliency maps to train our 3SD network. Specifically, we propose a contrastive learning-based training on multiple image patches for the classification task. We show the multi-patch classification with contrastive loss improves the quality of the CAM maps compared to naive classification on the entire image. Experiments on six benchmark datasets demonstrate that without any labels, our 3SD method outperforms all existing weakly supervised and unsupervised methods, and its performance is on par with the fully-supervised methods.

Recently, body part as an intuitive movement unit has received increasing attention in skeleton-based action. However, the part-level embedding is hard to be fully exploited, especially for fine-grained actions, as the body joints are aggregated into parts. To address this problem, we propose a novel transformer-based network (IIP-Transformer). Different from previous models that rely on specially designed partition method, our proposed IIPA mechanism which incorporates joint-level (intra-part) and part-level (inter-part) interactions simultaneously is the keypoint for IIP-Transformer to fully exploit part-level data, making considerable improvements in both coarse-grained and fine-grained action recognition. Ablation studies on three typical partition methods show that IIP-Transformer is a relatively general solution for part-level data and thus we could choose the simplest hand-craft partition embedding to significantly reduce computational complexity and model size. Besides, The proposed IIP-Transformer exceeds the state-of-the-art methods with much less computational complexity on NTU RGB+D, NTU RGB+D120 and NW-UCLA datasets.

In this paper, we propose a novel monocular ray-based 3D (Ray3D) absolute human pose estimation with calibrated camera. Accurate and generalizable absolute 3D human pose estimation from monocular 2D pose input is an ill-posed problem. To address this challenge, we convert the input from pixel space to 3D normalized rays. This conversion makes our approach robust to camera intrinsic parameter changes. To deal with the in-the-wild camera extrinsic parameter variations, Ray3D explicitly takes the camera extrinsic parameters as an input and jointly models the distribution between the 3D pose rays and camera extrinsic parameters. This novel network design is the key to the outstanding generalizability of Ray3D approach. To have a comprehensive understanding of how the camera intrinsic and extrinsic parameter variations affect the accuracy of absolute 3D key-point localization, we conduct in-depth systematic experiments on three single person 3D benchmarks as well as one synthetic benchmark. These experiments demonstrate that our method significantly outperforms existing state-of-the-art models. Our code and the synthetic dataset are available at https://github.com/YxZhxn/Ray3D.

The recent trend in multiple object tracking (MOT) is heading towards leveraging deep learning to boost the tracking performance. However, it is not trivial to solve the data-association problem in an end-to-end fashion. In this paper, we propose a novel proposal-based learnable framework, which models MOT as a proposal generation, proposal scoring and trajectory inference paradigm on an affinity graph. This framework is similar to the two-stage object detector Faster RCNN, and can solve the MOT problem in a data-driven way. For proposal generation, we propose an iterative graph clustering method to reduce the computational cost while maintaining the quality of the generated proposals. For proposal scoring, we deploy a trainable graph-convolutional-network (GCN) to learn the structural patterns of the generated proposals and rank them according to the estimated quality scores. For trajectory inference, a simple deoverlapping strategy is adopted to generate tracking output while complying with the constraints that no detection can be assigned to more than one track. We experimentally demonstrate that the proposed method achieves a clear performance improvement in both MOTA and IDF1 with respect to previous state-of-the-art on two public benchmarks. Our code is available at https://github.com/daip13/LPC_MOT.git.

Over the past three decades, a number of face recognition methods have been proposed in computer vision, and most of them use holistic face images for person identification. In many real-world scenarios especially some unconstrained environments, human faces might be occluded by other objects, and it is difficult to obtain fully holistic face images for recognition. To address this, we propose a new partial face recognition approach to recognize persons of interest from their partial faces. Given a pair of gallery image and probe face patch, we first detect keypoints and extract their local textural features. Then, we propose a robust point set matching method to discriminatively match these two extracted local feature sets, where both the textural information and geometrical information of local features are explicitly used for matching simultaneously. Finally, the similarity of two faces is converted as the distance between these two aligned feature sets. Experimental results on four public face data sets show the effectiveness of the proposed approach.

In this paper, we propose a new cascaded deep auto-encoder networks (CDAN) approach for face alignment. Our framework consists of a global exemplar-based deep auto-encoder network (GEDAN) and a series of localized deep auto-encoder networks (LDAN) in a cascaded fashion. The global network takes a low-resolution holistic facial image as input and generates a preliminary facial landmark configuration. The following localized networks sample pose-indexed local features around current landmark positions, and refine the landmark positions with increasingly higher image resolutions. Our network architectures are designed to achieve greater robustness against pose variations as well as higher landmark estimation accuracy. Experimental results on three datasets show that the proposed approach achieves superior alignment accuracy with real-time speed.

Over the past two decades, a number of face recognition methods have been proposed in the literature. Most of them use holistic face images to recognize people. However, human faces are easily occluded by other objects in many real-world scenarios and we have to recognize the person of interest from his/her partial faces. In this paper, we propose a new partial face recognition approach by using feature set matching, which is able to align partial face patches to holistic gallery faces automatically and is robust to occlusions and illumination changes. Given each gallery image and probe face patch, we first detect key points and extract their local features. Then, we propose a Metric Learned Extended Robust Point Matching (MLERPM) method to discriminatively match local feature sets of a pair of gallery and probe samples. Lastly, the similarity of two faces is converted as the distance between two feature sets. Experimental results on three public face databases are presented to show the effectiveness of the proposed approach.