Computer-vision researcher in visible–infrared object detection: semi-supervised learning and adversarial robustness.
2
papers here
1
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
—
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-09-21) 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—
Insufficient data: the output axis would measure publication volume and productivity — total paper counts, venue tiers, and yearly distribution. No such aggregate metrics are present in the current snapshot, so this axis is left unscored rather than estimated.
Impact—
Insufficient data: the impact axis would measure citation-based influence — citation counts, h-index, and the venues of citing works. These metrics are not available in the current snapshot, so no impact estimate is offered here.
Focus100/100
The focus score of 100 reflects an unusually narrow topical signature: the one field-classified paper in the catalog maps to computer vision, and both known works target visible–infrared object detection — one on semi-supervised oriented detection in weakly-aligned RGB–IR imagery, one attacking such detectors. The line is coherent, though it rests on a single classified paper, so it is directionally strong but statistically thin.
Collaboration59/100
Collaboration scores 59, with 18 distinct coauthors in the bibliometric snapshot. The recurring list is anchored by Qi Ming, who appears twice in the frequent-coauthor records, alongside Jiangmeng Li, Juan Fang, Xiaxin Yuan, Xudong Zhao, Zhanchao Huang, Shaoguang Huang, Aleksandra Pizurica, Yueqi Zhu, Guo Cheng, and Yongkang Zhang — a stable core team plus intermittent external partners. On both catalog papers the researcher sits in a middle authorship position rather than leading.
Activity—
Insufficient data: the activity axis would measure research cadence — publications per recent year, submission continuity, and ongoing projects. Both known papers date to 2026, too narrow a window to compute a reliable activity trend, so no score is assigned.
Reproducibility55/100
Reproducibility averages 55 over one assessed paper — the semi-supervised RGB–IR oriented detection study, whose Evolve readiness is rated medium. The review credits strong results but flags unmatched baselines and single-run peak numbers, which directly caps confidence in the headline gains. With only one paper evaluated, this score describes a single data point rather than an established replication track record.
Research summary
Jiahuan Zhou is a computer-vision researcher whose recorded work centers on visible–infrared (RGB–IR) object detection, with two 2026 arXiv preprints in the Evolve catalog. The first, on aligned consensus teaching for label-efficient oriented detection in weakly-aligned visible-infrared imagery, is reviewed by Evolve as the first semi-supervised framework for image-pair-level RGB–IR detection: a mean-teacher consensus recipe with pseudo labels and cross-modal alignment, delivering strong results whose headline gains are tempered by unmatched baselines and single-run peak numbers (reproducibility readiness medium, score 55). The second, 'Stealthy in Semantics, Antagonistic in Space', attacks visible–infrared detectors through object-level misalignment, extending the same line toward adversarial robustness. The profile shows a sharply focused signature — focus score 100, with the single field-classified paper assigned to computer vision — and a mid-range collaboration footprint of 18 distinct coauthors (collaboration score 59), with Qi Ming, Jiangmeng Li, and Aleksandra Pizurica among recurring partners. On both catalog papers Zhou holds a middle authorship position without corresponding-author role. Output, impact, and activity metrics are not yet available, so this is best read as an early, name-resolved bibliographic sketch of a multimodal, label-efficient detection specialist.
Collaboration pattern
Across the snapshot Jiahuan Zhou has 18 distinct coauthors (collaboration score 59). The frequent list is anchored by Qi Ming, who appears twice, with Jiangmeng Li, Juan Fang, Xiaxin Yuan, Xudong Zhao, Zhanchao Huang, Shaoguang Huang, Aleksandra Pizurica, Yueqi Zhu, Guo Cheng, and Yongkang Zhang as recurring partners — a stable core group plus intermittent external collaborators. On both catalog papers Zhou sits in a middle authorship position and is not the corresponding author.
Research keywords
Highlights
First semi-supervised framework for image-pair-level RGB–IR detection
Consensus teaching recipe for weakly-aligned imagery
Adversarial attack on visible–infrared detectors
The 2026 preprint 'Stealthy in Semantics, Antagonistic in Space' attacks visible-infrared object detectors via object-level misalignment, showing the same multimodal detection line also covers offensive robustness analysis. Zhou is a middle author and not the corresponding author; no Evolve review verdict or reproducibility rating exists yet for this paper.
Timeline
- 2026arXiv preprint: Aligned Consensus Teaching for Label-Efficient Oriented Object Detection in Weakly-Aligned Visible-Infrared Imagery — Evolve-reviewed as the first image-pair-level semi-supervised RGB–IR detection framework, reproducibility readiness medium.
- 2026arXiv preprint: Stealthy in Semantics, Antagonistic in Space — attacking visible-infrared object detectors via object-level misalignment, extending the recorded line into adversarial robustness.
- 2026Both catalog outputs land in the same year, making 2026 the recorded starting point of the visible–infrared detection line; output and activity trends cannot yet be assessed beyond this single-year window.
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 scores are computed from public bibliographic data (OpenAlex-based); citation counts, h-index, and venue-tier metrics were not available in this snapshot.
- No ORCID, institution, or country could be confirmed; 'Jiahuan Zhou' is a common name, so this profile is name-resolved from public bibliographic data and may merge or split same-name identities.
- Only two 2026 arXiv preprints are in the reviewed catalog; earlier work and journal or conference publications fall outside this data scope.
- Reproducibility (55) and focus (100) rest on a single reviewed, single field-classified paper, so these scores are indicative rather than established.