jetyingjia

I don't find any data_augmentation in this code. ex:scale、 shift、motion blur； which are very important in training! Am I miss some code? Thank you fot this project!

Table 6: Performance of Alpha-CLIP in region level captioning

1

Great work! I am confused with Tab .6 result, the performance is Alpha-CLIP with LLaVA-1.5 or fine-tune this model with vicuna-7b on these datasets(RefCOCOg or VG)？

Caption branch

2

Thanks for your great work！ In your project, the caption branch is trained only on VG data. This caption ability may be poor than the modal using large caption data...