Early-career gait recognition researcher at Beijing Normal University with three first-author 2026 arXiv papers
3
papers here
3
reviewed by Evolve
3
first-author
0
corresponding
—
works · OpenAlex
4
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-09-19) and cover the author's entire public record, not just the papers listed here. Reproducibility signal comes from Evolve's own reviews.
Research profile
Output—
This axis would measure total publication volume across the researcher's career. The snapshot provides only partial year-by-year tallies and no complete, deduplicated work count, so the axis is left unscored rather than estimated.
Impact—
This axis would measure citation-based influence such as aggregate and field-weighted citations. The snapshot records an h-index of 4 and an i10-index of 2 but lacks fuller citation coverage, so the axis is left unscored.
Focus8/100
The focus dimension scores 8/100. The underlying distribution is nonetheless concentrated: roughly two-thirds of indexed works fall in computer vision (gait recognition/biometrics) and one-third in computer vision overall, and the top OpenAlex topic is gait recognition and analysis (4), ahead of video surveillance and tracking methods (3) and human pose and action recognition (2), with single-count topics on falls prevention and context-aware activity recognition. The low score reflects spread across these adjacent clusters.
Collaboration51/100
The collaboration dimension scores 51/100. The recent three-year window covers 6 works with 12 distinct co-authors. The recurring core is compact and gait-focused: Saihui Hou and Yongzhen Huang appear three times each in the frequent-coauthor list, Xu Liu twice, with Chunshui Cao, Yunjie Peng, Zhiqiang He, and Junzhou Huang also present. This pattern suggests a stable in-house team rather than a wide cross-institution network.
Activity57/100
The activity dimension scores 57/100. Recorded year-by-year counts are 1 (2021), 3 (2024), and 3 (2025), with 6 works in the most recent three-year window; the Evolve catalog adds three first-author arXiv submissions dated 2026. The pattern shows a pause after 2021 followed by a sustained higher cadence from 2024 onward.
Reproducibility55/100
The reproducibility dimension scores 55/100, a mean across the 3 papers Evolve has scored. All three catalog papers carry a "medium" reproducibility readiness. Evolve's verdicts consistently flag the absence of statistical significance tests — thin margins in GaitMoE, and modest +0.5–4% gains at roughly double the parameters for vocabulary guidance — so reported improvements remain plausible but statistically unverified.
Research summary
Panjian Huang is an early-career computer-vision researcher at Beijing Normal University whose work centers on gait recognition as a biometric. The profile is name-resolved from OpenAlex bibliographic data linked to ORCID 0000-0002-9777-4365, and the metrics below reflect that snapshot. Three first-author 2026 arXiv papers anchor the catalog: a diffusion-inspired unified template for gait recognition reporting state-of-the-art results on six benchmarks; GaitMoE, which recasts occluded gait as action composition using temporal and action experts and introduces the OccGait dataset; and a vocabulary-guided approach that injects eight gait-phase text embeddings generated by large language models into gait networks, yielding consistent but modest gains (+0.5–4%) at roughly double the parameters. Recorded bibliometrics are correspondingly early-stage: h-index 4, i10-index 2, six works in the most recent three-year window, and year-by-year counts of 1 (2021), 3 (2024), and 3 (2025). Topical distribution is heavily gait-centric — gait recognition and analysis (4) leads the topic list, and about two-thirds of works sit in gait recognition/biometrics — with adjacent coverage of video surveillance and tracking, human pose and action recognition, and falls prevention. Twelve distinct co-authors appear in the recent window, with Saihui Hou and Yongzhen Huang the most frequent collaborators. Evolve's review verdicts repeatedly flag missing statistical significance tests, so headline claims should be read as promising but not yet statistically validated.
Collaboration pattern
Twelve distinct co-authors appear across the recent three-year window of six works. The collaboration list is anchored by a compact gait-focused group: Saihui Hou and Yongzhen Huang each recur three times, Xu Liu twice, alongside Chunshui Cao, Yunjie Peng, Zhiqiang He, and Junzhou Huang. The radar collaboration dimension scores 51/100, consistent with a small stable core team rather than a broad cross-institution network.
Research keywords
Highlights
Unified gait template via diffusion reports SOTA on six benchmarks
GaitMoE recasts occluded gait as action composition
LLM gait-phase vocabulary yields consistent but modest gains
Timeline
- 2021Earliest year recorded in the OpenAlex snapshot, with 1 indexed work.
- 20243 indexed works recorded in 2024, marking a clear step up in publication cadence after the 2021 entry.
- 2025Another 3 indexed works recorded in 2025, sustaining the higher output level.
- 2026Three first-author arXiv preprints enter the Evolve catalog: a diffusion-inspired unified gait template, the GaitMoE mixture-of-experts model for occluded gait, and vocabulary-guided gait recognition.
Papers in the Evolve catalog
Frequent co-authors
About this profile
This profile is AI-structured by Evolve.Pub from public bibliographic data (OpenAlex) and Evolve's own paper reviews. It is not written or endorsed by the researcher unless the profile is claimed. Radar dimensions are computed from real bibliometrics with fixed formulas — raw figures are shown alongside; missing dimensions mean insufficient data, never zero. Errors? Claim the profile to correct or hide it.
- Bibliometric metrics come from a single OpenAlex snapshot (identity: openalex) linked to ORCID 0000-0002-9777-4365; the output and impact radar axes are unscored for insufficient data.
- Panjian Huang is not a unique name; this profile is name-resolved from public bibliographic data, so some indexed works could belong to different individuals sharing the name.
- All three catalog papers are 2026 arXiv preprints; peer-review status and final publication venues are not reflected in this snapshot.
- Reproducibility scoring covers only the 3 papers Evolve has reviewed; broader code and artifact release behavior is not assessed.