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

This is the repository for our paper ModPrompt: Visual (Mod)ality (Prompt) for Adapting Vision-Language Object Detectors 🔗 by Heitor Rapela Medeiros, Atif Belal, Srikanth Muralidharan, Eric Granger and Marco Pedersoli.

News

• Paper is under review, code will be released soon.
• Arxiv: https://arxiv.org/pdf/2412.00622
• If you find any problem or have any questions, please feel free to contact us!

ModPrompt + Task Residuals

Benchmarking

References

- Thanks to the great open-source community that provided good libraries.
- This code is based on MMDET, YOLO-World, Grounding DINO and Visual Prompt.