Deep Intra-Image Contrastive Learning for Weakly Supervised One-Step Person Search

Abstract

Weakly supervised person search aims to perform joint pedestrian detection and re-identification (re-id) with only person bounding-box annotations. Recently, the idea of contrastive learning is initially applied to weakly supervised person search, where two common contrast strategies are memorybased contrast and intra-image contrast. We argue that current intra-image contrast is shallow, which suffers from spatiallevel and occlusion-level variance. In this paper, we present a novel deep intra-image contrastive learning using a Siamese network. Two key modules are spatial-invariant contrast (SIC) and occlusion-invariant contrast (OIC). SIC performs many-toone contrasts between two branches of Siamese network and dense prediction contrasts in one branch of Siamese network. With these many-to-one and dense contrasts, SIC tends to learn discriminative scale-invariant and location-invariant features to solve spatial-level variance. OIC enhances feature consistency with the masking strategy to learn occlusion-invariant features. Extensive experiments are performed on two person search datasets CUHK-SYSU and PRW, respectively. Our method achieves a state-of-the-art performance among weakly supervised one-step person search approaches. We hope that our simple intra-image contrastive learning can provide more paradigms on weakly supervised person search. The source code is available at https://github.com/jiabeiwangTJU/DICL.

View-target relation-guided unsupervised 2D image-based 3D model retrieval via transformer

“…To begin with, for feature representation of both 2D images and 3D models, a better backbone is always encouraged, which draws our attention to the trendy vision transformers (ViT) recently. It has proved to be a success in many relative computer vision and natural language processing (NLP) such as video event detection [16], pedestrian detection [17], person search [18,19], and text classification [20]. ViT takes the image patch or word embedding as a sequence of tokens, and applies the self-attention mechanism to capture the internal relationships thus obtaining strong feature representation connected with downstream tasks.…”