Background subtraction is a classic video processing task pervading in numerous visual applications such as video surveillance and traffic monitoring. Given the diversity and variability of real application scenes, an ideal background subtraction model should be robust to various scenarios. Even though deep-learning approaches have demonstrated unprecedented improvements, they often fail to generalize to unseen scenarios, thereby less suitable for extensive deployment. In this work, we propose to tackle cross-scene background subtraction via a two-phase framework that includes meta-knowledge learning and domain adaptation. Specifically, as we observe that meta-knowledge (i.e., scene-independent common knowledge) is the cornerstone for generalizing to unseen scenes, we draw on traditional frame differencing algorithms and design a deep difference network (DDN) to encode meta-knowledge especially temporal change knowledge from various cross-scene data (source domain) without intermittent foreground motion pattern. In addition, we explore a self-training domain adaptation strategy based on iterative evolution. With iteratively updated pseudo-labels, the DDN is continuously fine-tuned and evolves progressively toward unseen scenes (target domain) in an unsupervised fashion. Our framework could be easily deployed on unseen scenes without relying on their annotations. As evidenced by our experiments on the CDnet2014 dataset, it brings a significant improvement to background subtraction. Our method has a favorable processing speed (70 fps) and outperforms the best unsupervised algorithm and top supervised algorithm designed for unseen scenes by 9% and 3%, respectively.

Monocular 3D human pose estimation is used to calculate a 3D human pose from monocular images or videos. It still faces some challenges due to the lack of depth information. Traditional methods have tried to disambiguate it by building a pose dictionary or using temporal information, but these methods are too slow for real-time application. In this paper, we propose a real-time method named G2O-pose, which has a high running speed without affecting the accuracy so much. In our work, we regard the 3D human pose as a graph, and solve the problem by general graph optimization (G2O) under multiple constraints. The constraints are implemented by algorithms including 3D bone proportion recovery, human orientation classification and reverse joint correction and suppression. When the depth of the human body does not change much, our method outperforms the previous non-deep learning methods in terms of running speed, with only a slight decrease in accuracy.

Multi-view 3D reconstruction technology based on deep learning is developing rapidly. Unsupervised learning has become a research hotspot because it does not need ground truth labels. The current unsupervised method mainly uses 3DCNN to regularize the cost volume to regression image depth. This approach results in high memory requirements and long computing time. In this paper, we propose an end-to-end unsupervised multi-view 3D reconstruction network framework based on PatchMatch, Unsup_patchmatchnet. It dramatically reduces memory requirements and computing time. We propose a feature point consistency loss function. We incorporate various self-supervised signals such as photometric consistency loss and semantic consistency loss into the loss function. At the same time, we propose a high-resolution loss method. This improves the reconstruction of high-resolution images. The experiment proves that the memory usage of the network is reduced by 80% and the running time is reduced by more than 50% compared with the network using 3DCNN method. The overall error of reconstructed 3D point cloud is only 0.501 mm. It is superior to most current unsupervised multi-view 3D reconstruction networks. Then, we test on different data sets and verify that the network has good generalization.