tart All work

COMPUTER VISION / RESEARCH

Expression-Guided Person Segmentation

A fixed-camera image and a natural-language description go in; a pixel-level person mask—or an explicit no-match result—comes out.

KMITL Project 1 · Team research

My contribution

Contributed to experiment implementation, evidence tooling, comparative evaluation, and technical reporting across the project workflow.

Download CV (English)

Try it · mini version

See what it can do

MOTS20 comparison of the human reference, original CoHD result, Scale704 result, and Scale768 result
Latest progress-report evidence: the same MOTS20 frame compared across the reference and three inference settings.

KMITL · CE69-27 Project 1 · Work in progress

What this project does

The research asks whether language-guided segmentation can identify a described person in fixed-camera surveillance imagery and return a pixel mask, including an empty mask when no person matches. LISA-7B-v1, SSP-SAM-224, and CoHD-Tiny are compared under shared evaluation protocols.

Inside the project

  • On the 90-image gRefCOCO slice, CoHD-Tiny reached 78.01% positive-target mean IoU and ran about 21.6× faster than LISA in the reported setup.
  • On the locked 120-frame confirmation set, Scale704 with a guard raised MOTS coverage recall from 62.39% to 77.32%.
  • On VIRAT, matched people increased from 10 to 48 of 635 while false-positive components fell from 150 to 120; mean time changed from 0.6541 to 0.5934 seconds per image.

Skills & practical experience

Practical experience with referring segmentation, vision-language models, controlled evaluation, ablation-style comparisons, evidence tracking, and reporting results with explicit limits.

PyTorchComputer visionVision-language modelsReferring segmentationEvaluation protocolsCCTV analysisResearch reporting

Figures and measurements come from the latest 30% progress report and its generated thesis assets. The current evidence covers two fixed daytime viewpoints; many distant or small people are still missed.

Project context

Active academic research on two fixed daytime CCTV viewpoints. Results show progress on the evaluated samples; they do not establish unseen-camera or broad real-world generalization.

CCTV image + expressionModel comparisonCandidate maskAutomated checksMask or no match

Project media

Research framework for expression-guided person segmentation in fixed-camera surveillance imagery
Research framework: shared inputs, baseline and candidate models, verification, and final mask output.
VIRAT surveillance comparison for original CoHD, Scale704, and Scale768 settings
Real VIRAT comparison used in the latest progress report.
Evidence portal displaying project experiments and generated artifacts
The project evidence portal keeps experiment outputs reviewable for reporting.