A new survey paper published on arXiv provides a comprehensive review of 3D skeleton-based person re-identification (SRID) methods. The paper categorizes existing techniques into hand-crafted, sequence-based, and graph-based modeling approaches. It also details various learning paradigms, including supervised, self-supervised, and unsupervised methods, and evaluates the performance of state-of-the-art SRID techniques on benchmark datasets. The authors highlight current challenges and future research directions, as well as interdisciplinary applications. AI
RANK_REASON The cluster contains an academic survey paper detailing advances in a specific AI research area. [lever_c_demoted from research: ic=1 ai=1.0]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →