Qi Ming

—

Unclaimed · AI-structured profileOpenAlex

Emerging computer-vision researcher in cross-view geo-localization and visible-infrared object detection

papers here

reviewed by Evolve

first-author

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—

The underlying data do not yet include publication counts, venue records, or indexing coverage for this researcher, so output is not scored. This axis would measure total and per-year research production once those records are linked.

Impact—

Citation counts, h-index, and influence indicators are not available in the current snapshot, so impact is not scored. Once citation records are linked, this axis would measure how widely the reviewed papers are cited and taken up.

Focus0/100

The focus score is 0/100, the lowest concentration band, derived from a field distribution split evenly (0.5/0.5) between cross-view geo-localization / remote sensing and computer vision more broadly. The three reviewed papers indeed span geo-localization, semi-supervised visible-infrared detection, and adversarial attack. With only three papers assessed, this spread may overstate dispersion; the score reflects measured topical breadth, not research quality.

Collaboration64/100

The collaboration score is 64/100, computed from 24 distinct coauthors, 12 of whom recur frequently — including Xudong Zhao, Yuhan Zhang, Jiahuan Zhou, Jiangmeng Li, Juan Fang, and Zhu Han. Across the three reviewed 2026 papers, authorship position varies: first author on the semi-supervised RGB-IR detection paper and middle-listed on the other two, with no corresponding-author role recorded, indicating active but not yet leading team-level coordination.

Activity—

Recent publishing cadence and ongoing research momentum cannot be computed from the available data, so this axis is unscored. It would measure year-over-year output rhythm and whether new work continues to appear in recent periods.

Reproducibility55/100

Reproducibility scores 55/100, a mean over two assessed papers (reproPapers = 2), both rated medium reproducibility readiness. For the first-author paper, Evolve's review notes strong results but unmatched baselines and single-run peaks; for the MoE geo-localization paper, VLM reranking pipeline costs are unaccounted. These factors temper but do not negate replication prospects.

Research summary

Qi Ming is a computer-vision researcher whose reviewed work to date concentrates on two adjacent problems: cross-view object geo-localization for UAV and street-view imagery, and label-efficient object detection in visible-infrared (RGB-IR) imagery. Evolve's catalog contains three 2026 arXiv papers. As first author of 'Aligned Consensus Teaching for Label-Efficient Oriented Object Detection in Weakly-Aligned Visible-Infrared Imagery' (2026), Qi proposes what Evolve's review describes as the first semi-supervised framework for image-pair-level RGB-IR detection, built on mean-teacher self-training, pseudo labels, and cross-modal alignment; the review notes that unmatched baselines and single-run peaks temper headline gains. As a middle-listed co-author, Qi contributed to 'Multi-View Mixture-of-Experts with Vision-Language Reranking for Cross-View Object Geo-Localization' (2026), which unifies UAV and street-view CVOGL in one MoE model with VLM reranking and SAM ellipse priors, and to 'Stealthy in Semantics, Antagonistic in Space' (2026), an object-level misalignment attack on visible-infrared detectors. Snapshot bibliometrics give a collaboration score of 64/100 with 24 distinct coauthors (12 recurring, including Xudong Zhao, Yuhan Zhang, Jiahuan Zhou, and Jiangmeng Li) and a reproducibility score of 55/100 averaged over two assessed papers, both rated medium readiness; the field distribution splits evenly between cross-view geo-localization/remote sensing and computer vision, giving a focus score of 0/100. Output, impact, and activity axes lack sufficient data and are not scored. 'Qi Ming' is a common name, so this profile is name-resolved from public bibliographic data.

Collaboration pattern

Bibliometric data show 24 distinct coauthors, with 12 recurring collaborators including Xudong Zhao, Yuhan Zhang, Jiahuan Zhou, Jiangmeng Li, Juan Fang, and Zhu Han. Authorship positions in the reviewed set are mixed: first author on the semi-supervised RGB-IR detection paper, middle-listed on the other two, and no corresponding-author role recorded. The collaboration score of 64/100 reflects a moderately sized, stable network centered on visible-infrared detection and geo-localization work.

Research keywords

cross-view object geo-localizationmixture-of-expertsvision-language rerankingSAM (Segment Anything) priorssemi-supervised object detectionvisible-infrared fusionoriented object detectionmean-teacher self-trainingpseudo labelscross-modal alignmentadversarial attack on detectorsUAV / drone navigation

Highlights

First-author semi-supervised framework for weakly-aligned RGB-IR detection
As first author of 'Aligned Consensus Teaching for Label-Efficient Oriented Object Detection in Weakly-Aligned Visible-Infrared Imagery' (2026), Qi proposes a framework combining mean-teacher self-training, pseudo labels, and cross-modal alignment; Evolve's review calls it the first semi-supervised framework for image-pair-level RGB-IR detection, while noting unmatched baselines and single-run peaks temper headline gains.
Co-author on multi-view MoE geo-localization with vision-language reranking
In 'Multi-View Mixture-of-Experts with Vision-Language Reranking for Cross-View Object Geo-Localization' (2026, middle-listed author), the team unifies UAV and street-view CVOGL in one MoE model with VLM reranking and SAM ellipse priors; Evolve notes strong UAV gains while street-view SOTA is only partial and VLM pipeline costs are unaccounted.
Co-author on adversarial attack against visible-infrared detectors
'Stealthy in Semantics, Antagonistic in Space: Attacking Visible-Infrared Object Detectors via Object-Level Misalignment' (2026, middle-listed author) studies object-level misalignment attacks on RGB-IR detectors; Evolve has not yet published a verdict, keywords, or reproducibility assessment for this paper.

Timeline

2026First-authored 'Aligned Consensus Teaching for Label-Efficient Oriented Object Detection in Weakly-Aligned Visible-Infrared Imagery' (arXiv).
2026Co-authored 'Stealthy in Semantics, Antagonistic in Space: Attacking Visible-Infrared Object Detectors via Object-Level Misalignment' (arXiv).
2026Co-authored 'Multi-View Mixture-of-Experts with Vision-Language Reranking for Cross-View Object Geo-Localization' (arXiv).

Papers in the Evolve catalog

Frequent co-authors

Xudong Zhao×2 Yuhan Zhang×2 Jiahuan Zhou×2 Jiangmeng Li×2 Juan Fang×2 Xuyu Fan×1 Zhu Han×1 Liuqian Wang×1 Siyuan Cao×1 Xiaohan Zhang×1 Mingjing Zhao×1 Xiaxin Yuan×1

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 come from a public bibliometric snapshot (OpenAlex-based); output, impact, and activity axes lack sufficient data and are not scored.
Only three 2026 arXiv preprints reviewed by Evolve are covered; they may not represent the full publication record, and preprint status means peer review may be pending.
No institution, country, or ORCID is available; 'Qi Ming' is a common name, so this profile is name-resolved from public bibliographic data and may merge or split identities.
The reproducibility score (55/100) is averaged over only two assessed papers, both rated medium readiness.