One‐shot video‐based person re‐identification with variance subsampling algorithm

Abstract

Previous works propose the distance-based sampling for unlabeled datapoints to address the few-shot person re-identification task, however, many selected samples may be assigned with wrong labels due to poor feature quality in these works, which negatively affects the learning procedure. In this article, we propose a novel sampling strategy to improve the quality of assigned pseudo-labels, thus promoting the final performance. To illustrate, we first propose the concept of variance confidence to measure the credibility of pseudo-labels, then we apply a novel variance subsampling algorithm to improve the accuracy of the selected sample labels. Our method combines distance confidence and variance confidence as a two-round sampling criterion. Meanwhile, a variation decay strategy is used in our sampling process in combination with the actual distribution of features. We evaluate our approach on two publicly available datasets, MARS and DukeMTMC-VideoReID, and achieve state-of-the-art one-shot performance.

Multi-Level Fusion Temporal–Spatial Co-Attention for Video-Based Person Re-Identification

“…However, these methods require additional information, usually through the different collections, which is time-consuming and laborious. There are also some methods that convert video sequences into image pairs, and use image-based person re-recognition models [ 14 , 15 , 16 , 17 ] to mine video sequence frame features. There are also some 2D or 3D convolution methods based on recurrent neural networks (RNNs) [ 18 ] to deal with the problem of feature extraction of video sequence frames.…”