First-author researcher bridging real-to-sim articulated-object learning and self-supervised surgical video tracking.
2
papers here
2
reviewed by Evolve
2
first-author
0
corresponding
—
works · OpenAlex
—
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-10-15) 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 is not scored: the catalogued record covers only 2 arXiv preprints from 2026, with no publication counts, venue history, or year-by-year output data yet. It would measure total research productivity and how consistently that output is sustained over time.
Impact—
No impact score is available: citation counts, h-index-type measures, and evidence of downstream adoption are absent from the current data, which spans only 2 recent preprints. This axis would measure influence on follow-up research and on practice.
Focus0/100
The focus axis scores 0, grounded in an even field distribution: 50% medical imaging / surgical computer vision and 50% computer vision / robotics (embodied AI). Each of the two catalogued 2026 preprints sits in a different field, so the small sample cannot yet distinguish deliberate cross-field method transfer — point tracking serves both — from an unfocused early spread.
Collaboration32/100
Collaboration scores 32, based on a compact pool of 4 distinct co-authors across the catalogued work. Recurring collaborators include Zijian Wu, Mehran Armand, Septimiu Salcudean, and Homanga Bharadhwaj. On both papers this researcher is first author and not corresponding, consistent with an early-career pattern of leading experimental work inside a small stable team rather than an extended network.
Activity—
Activity is unscored: the dataset lacks per-year publication volume, submission cadence, and timestamps beyond two 2026 entries. It would measure ongoing research pace and engagement across consecutive years.
Reproducibility38/100
Reproducibility averages 38 across the 2 papers with reproducibility assessments. Track, Articulate, Act shows medium readiness, aided by MuJoCo-replayable articulated outputs, though its real-video evidence remains qualitative. S3-Tracker shows low readiness, and its review notes accuracy trails MFT on both metrics with the advantage limited to latency. The sample is small, so this score is preliminary.
Research summary
Jiaming Zhang is an early-stage researcher whose catalogued output comprises two first-author arXiv preprints, both dated 2026, split evenly between surgical computer vision and embodied-AI robotics. The first, 'Track, Articulate, Act,' converts casual monocular RGB videos of human interaction into articulated object models replayable in MuJoCo, using 3D point tracks and joint estimation; Evolve's review credits its strength in simulation while noting that real-video evidence stays qualitative and end-to-end robustness is untested. The second, 'S3-Tracker,' applies self-supervised contrastive random walks to endoscopic tissue tracking with visibility estimation; the review finds it trails the semi-supervised MFT tracker on both accuracy metrics in Table I, winning only on latency. Bibliometrically the record is thin: 4 distinct co-authors — among them Zijian Wu, Mehran Armand, Septimiu Salcudean, and Homanga Bharadhwaj — a mean reproducibility score of 38 across the 2 assessed papers, and an even 50/50 field split. No institution, country, or ORCID is on record, and the name is common, so this profile is name-resolved from public bibliographic data. Output, impact, and activity axes are not yet scored.
Collaboration pattern
Collaboration scores 32 from a compact pool of 4 distinct co-authors. Recurring names include Zijian Wu, Mehran Armand, Septimiu Salcudean, and Homanga Bharadhwaj, spanning both catalogued fields. On both 2026 preprints this researcher is first author and not corresponding — a pattern typical of early-career, lead experimental contributions inside a small stable team rather than a wide collaboration network.
Research keywords
Highlights
From casual RGB videos to MuJoCo-replayable articulated objects
Self-supervised surgical tissue tracking via contrastive random walks
Timeline
- 2026First-author arXiv preprint 'Track, Articulate, Act' proposes generating articulation from casual human videos.
- 2026First-author arXiv preprint 'S3-Tracker' introduces self-supervised surgical tissue tracking with contrastive random walks.
- 2026Both catalogued works appear in the same year, splitting output evenly between embodied-AI computer vision and surgical computer vision.
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.
- Name-resolved from public bibliographic data: 'Jiaming Zhang' is a common name, and no institution, country, or ORCID is on record.
- All metrics come from a very small sample: 2 arXiv preprints, both dated 2026; scores should be read as preliminary.
- Output, impact, and activity axes are unscored because the underlying data is insufficient.
- No citation counts or peer-reviewed venue data are in scope; the reproducibility score reflects Evolve's review of these 2 papers.