Inquiry on Methodology Used in SenseTime/deformable-detr Model
#4
by
hchcsuim
- opened
I noticed three methods mentioned in the Deformable DETR paper: Deformable DETR, Deformable DETR with iterative bounding box refinement, and two-stage Deformable DETR. Could you clarify which of these methods is implemented in the SenseTime/deformable-detr model?
Hi,
All checkpoints can be found here: https://huggingface.co/models?search=sensetime/deformable-detr. This model corresponds to the single scale model released at https://github.com/fundamentalvision/Deformable-DETR?tab=readme-ov-file#main-results.