ABOUT THIS COLLECTION ×
A companion to the survey. This library brings together the paper summaries used to organize Modern Gait Recognition: A Survey from Deep Learning to Foundation Models
Reading the summaries Core Motivation describes the problem being addressed. Key Idea captures the proposed insight. Main Techniques identifies how the idea is implemented. These concise editorial summaries are not quotations from the papers.
Expand Abstract to read the paper abstract. Entries labeled Abstract · Summary contain a concise editorial summary of the linked publication. Original abstracts and editorial summaries are distinguished explicitly.
Exploring the companion Benchmarks connects within- and cross-domain rankings to the experimental records in Tables III–IV. Statistics reproduces Table II and Figures 5–6 from one publication list. Research map connects papers to the five methodological dimensions and the discussion in Sections V–VI; links carry indexed passages or clearly labeled manuscript summaries. The progress guide follows Table V and separates demonstrated gains from remaining limits; synthetic data and scaling provide cross-cutting reading routes. Healthcare asks which changes should remain visible, while agents bring together long-video understanding and robotics as prospective applications. Projects & competitions provides searchable research programs and competition editions with official sources, evaluation settings and downloadable citations. The repository contains a documented, validated workflow for adding papers and releasing new snapshots.
Reading the word cloud Each term is counted at most once per matching paper, using the selected summary fields. Common function words and generic writing terms are excluded. The cloud updates with the search, year and venue filters, and includes only the 2020–2025 statistics membership from the specified venues. Its sizes describe this collection, not citation impact or the complete field.
Conference and journal versions remain separate when their publication records differ. Withdrawn records are marked and excluded from the word cloud. The 2026 collection is still partial and excluded from the frozen statistics. Venue scope: CVPR, ICCV, ECCV, WACV, BMVC, ACCV, IJCB, AAAI, ACM MM; TPAMI, TIFS, TIP, TNNLS, TCSVT, TBIOM, TMM, IJCV, PR. PRL, workshops, NeurIPS and preprints remain outside those statistics; result sources for the benchmark tables are retained separately.
