Z
Zimo Liu
Researcher at Dalian University of Technology
Publications - 6
Citations - 422
Zimo Liu is an academic researcher from Dalian University of Technology. The author has contributed to research in topics: Visibility (geometry) & Computer science. The author has an hindex of 4, co-authored 4 publications receiving 232 citations.
Papers
More filters
Proceedings ArticleDOI
Stepwise Metric Promotion for Unsupervised Video Person Re-identification
Zimo Liu,Dong Wang,Huchuan Lu +2 more
TL;DR: This paper proposes a stepwise metric promotion approach to estimate the identities of training tracklets, which iterates between cross-camera tracklet association and feature learning, and can eliminate the hard negative label matches.
Proceedings ArticleDOI
Pose-Guided Visible Part Matching for Occluded Person ReID
TL;DR: A Pose-guided Visible Part Matching (PVPM) method that jointly learns the discriminative features with pose-guided attention and self-mines the part visibility in an end-to-end framework that achieves competitive performance to state-of-the-art methods.
Proceedings ArticleDOI
Deep Reinforcement Active Learning for Human-in-the-Loop Person Re-Identification
TL;DR: An alternative reinforcement learning based human-in-the-loop model which releases the restriction of pre-labelling and keeps model upgrading with progressively collected data is proposed to minimize human annotation efforts while maximizing Re-ID performance.
Posted Content
Pose-guided Visible Part Matching for Occluded Person ReID
TL;DR: Zhang et al. as mentioned in this paper proposed a Pose-Guided Visible Part Matching (PVPM) method that jointly learns the discriminative features with pose-guided attention and self-mines the part visibility in an end-to-end framework.
Journal ArticleDOI
Towards Mitigating the Problem of Insufficient and Ambiguous Supervision in Online Crowdsourcing Annotation
TL;DR: This paper investigates a more general and broadly applicable learning problem, i.e. semi-supervised partial label learning, and proposes a novel method based on pseudo-labeling and contrastive learning, which consistently outperforms all comparing methods by a significant margin and set up the first state-of-the-art performance for semi- supervised partiallabel learning on image benchmarks.