Occluded person re-identification (Re-ID) focuses on addressing the occlusion problem when retrieving the person of interest across non-overlapping cameras. With the increasing demand for intelligent video surveillance and the application of person Re-ID technology, the real-world occlusion problem draws considerable interest from researchers. Although a large number of occluded person Re-ID methods have been proposed, there are few surveys that focus on occlusion. To fill this gap and help boost future research, this paper provides a systematic survey of occluded person Re-ID. In this paper, we review recent deep learning-based occluded person Re-ID researches. Firstly, we summarize the main issues caused by occlusion as four groups, i.e., position misalignment, scale misalignment, noisy information, and missing information. Secondly, we categorize existing methods into six solution groups, i.e., matching, image transformation, multi-scale features, attention mechanism, auxiliary information, and contextual recovery. We also discuss the characteristics of each approach, as well as the issues they addressed. Furthermore, we present the performance comparison of recent occluded person Re-ID methods on four public datasets: Partial-ReID, Partial-iLIDS, Occluded-ReID, and Occluded-DukeMTMC. We conclude the study with thoughts into promising future research directions.

With the widespread adoption of Cloud computing system, more and more application services based on cluster computing adopt this new computing paradigm. Live snapshot technology is usually used to ensure the reliability of data and high availability of the running virtual machines. This technology is essential to bring abilities of fault recovery and rollback to virtual clusters. Existing virtual machine live snapshot technologies are not enough to cope with the following challenges induced by taking live snapshots for a virtual cluster. One is keeping communication status among virtual machines belonging to a virtual cluster, and the other is minimizing the downtime of virtual machines when taking snapshot for a virtual cluster consisting of a large number of virtual machines. To address both challenges, this paper proposes a virtual cluster live snapshot framework which aims to preserve communication status among virtual machines belonging to a virtual cluster and to minimize the virtual machines downtime and performance overhead of live snapshot. Our experiments show that there is no lost package during taking live snapshot or restoring live snapshot, and that the average downtime of eight virtual machines in a virtual cluster is just 3% longer than that of individual virtual machine during taking live snapshot.

Gait recognition captures gait patterns from the walking sequence of an individual for identification. Most existing gait recognition methods learn features from silhouettes or skeletons for the robustness to clothing, carrying, and other exterior factors. The combination of the two data modalities, however, is not fully exploited. Previous multimodal gait recognition methods mainly employ the skeleton to assist the local feature extraction where the intrinsic discrimination of the skeleton data is ignored. This paper proposes a simple yet effective Bimodal Fusion (BiFusion) network which mines discriminative gait patterns in skeletons and integrates with silhouette representations to learn rich features for identification. Particularly, the inherent hierarchical semantics of body joints in a skeleton is leveraged to design a novel Multi-Scale Gait Graph (MSGG) network for the feature extraction of skeletons. Extensive experiments on CASIA-B and OUMVLP demonstrate both the superiority of the proposed MSGG network in modeling skeletons and the effectiveness of the bimodal fusion for gait recognition. Under the most challenging condition of walking in different clothes on CASIA-B, our method achieves the rank-1 accuracy of 92.1%.