CASIA-IVA-Lab/DPT
DPT: Deformable Patch-based Transformer for Visual Recognition (ACM MM2021)
PythonApache-2.0
Issues
- 13
- 1
Visualization of boxes
#19 opened by o0t1ng0o - 3
About the patch size
#18 opened by Jarviswang94 - 7
Could you kindly share the code to visualize the learned patches like Fig.6 in your paper ?
#10 opened by MaFuyan - 3
- 0
- 2
- 0
How 同
#14 opened by JusticeLin - 0
How to use DPT in ViT?
#15 opened by JusticeLin - 0
- 5
How to use DPT in DETR
#11 opened by lingblessing - 10
Use when the H and W are different ...
#9 opened by lingblessing - 5
- 4
- 2
Can it be used in a 3D scene?
#6 opened by zhixiaoqiang-haha - 1
creat model: dpt_small failed
#5 opened by Wyntor - 2
input image=112*112
#4 opened by eeric - 0
- 1