Thank you for sharing the impressive work! From the paper, I found that NAVER utilizes `GLaMM` for segmentation tasks (such as RefCOCOg). <img width="497" height="75" alt="Image" src="https://github.com/user-attachments/assets/ae15a764-3c1d-4269-9bc4-ffc600a073b9" /> However, I could't find `GLaMM` model in this repo, and the `main.py` script supports only referring object detection tasks. Could you update the code to include the segmentation task?
Thank you for sharing the impressive work!
From the paper, I found that NAVER utilizes
GLaMMfor segmentation tasks (such as RefCOCOg).However, I could't find
GLaMMmodel in this repo, and themain.pyscript supports only referring object detection tasks.Could you update the code to include the segmentation task?