Since the person re-identification task often suffers from the problem of pose changes and occlusions, some attentive local features are often suppressed when training CNNs. In this paper, we propose the Batch DropBlock (BDB) Network which is a two branch network composed of a conventional ResNet-50 as the global branch and a feature dropping branch. The global branch encodes the global salient representations. Meanwhile, the feature dropping branch consists of an attentive feature learning module called Batch DropBlock, which randomly drops the same region of all input feature maps in a batch to reinforce the attentive feature learning of local regions. The network then concatenates features from both branches and provides a more comprehensive and spatially distributed feature representation. Albeit simple, our method achieves state-of-the-art on person re-identification and it is also applicable to general metric learning tasks. For instance, we achieve 76.4% Rank-1 accuracy on the CUHK03-Detect dataset and 83.0% Recall-1 score on the Stanford Online Products dataset, outperforming the existing works by a large margin (more than 6%).

Unsupervised person re-identification (re-ID) attracts increasing attention due to its practical applications in industry. State-of-the-art unsupervised re-ID methods train the neural networks using a memory-based non-parametric softmax loss. They store the pre-computed instance feature vectors inside the memory, assign pseudo labels to them using clustering algorithm, and compare the query instances to the cluster using a form of contrastive loss. During training, the instance feature vectors are updated. However, due to the varying cluster size, the updating progress for each cluster is inconsistent. To solve this problem, we present Cluster Contrast which stores feature vectors and computes contrast loss in the cluster level. We demonstrate that the inconsistency problem for cluster feature representation can be solved by the cluster-level memory dictionary. By straightforwardly applying Cluster Contrast to a standard unsupervised re-ID pipeline, it achieves considerable improvements of 9.5%, 7.5%, 6.6% compared to state-ofthe-art purely unsupervised re-ID methods and 5.1%, 4.0%, 6.5% mAP compared to the state-of-the-art unsupervised domain adaptation re-ID methods on the Market, Duke, and MSMT17 datasets.

No abstract

No abstract

No abstract

The new dimensional deformation approach is proposed to generate higher dimensional analogues of integrable systems. An arbitrary (K+1)-dimensional integrable Korteweg-de Vries (KdV) system, as an example, exhibiting symmetry, is illustrated to arise from a reconstructed deformation procedure, starting with a general symmetry integrable (1+1)-dimensional dark KdV system and its conservation laws. Physically, the dark equation systems may be related to dark matter physics. To describe nonlinear physics, both linear and nonlinear dispersions should be considered. In the original lower dimensional integrable systems, only liner or nonlinear dispersion is included. The deformation algorithm naturally makes the linear dispersion and nonlinear dispersion are included in the same model.

No abstract

No abstract