[ICCV2025] ModPrompt: Visual Modality Prompt for Adapting Vision-Language Object Detectors

adaptation grounding-dino objectdetection visual-prompt vpt yolo-world yolo-world-v2
1 Open Issue Need Help Last updated: Jul 10, 2025

Open Issues Need Help

View All on GitHub
Releasing of the code about 2 months ago

AI Summary: Release the code for the ModPrompt project, a visual modality prompt for adapting vision-language object detectors, as a first version. The code is based on existing libraries (MMDET, YOLO-World, Grounding DINO, and Visual Prompt). Future improvements include cleaning the code, adding more details on how to run it, and providing pretrained weights.

Complexity: 4/5
good first issue

[ICCV2025] ModPrompt: Visual Modality Prompt for Adapting Vision-Language Object Detectors

Python
#adaptation#grounding-dino#objectdetection#visual-prompt#vpt#yolo-world#yolo-world-v2