Phyllis
Phyllis
Hi~ Thanks for providing the tensorflow version of the show-attend-and-tell model! I'm trying to do experiments by using this code. Could u tell me the best performance it can achieve...
The fast-rcnn will generate bunches of bounding boxes? How do you decide the amount of bouding boxes for each image in the proposal.pkl file?
Hi. Thanks for release the evaluation function. I've written one while that can not achive the published result when k=70. Could you release the evaluation when k=70? Thanks sooooooo much....
Hi~ Thanks for your excellent work! Could the current codebase support BEV or 3D tracking? And if not, how could I modify it to achieve this? Looking forward to your...
Such as anchor-free detector (centernet), transformer-based detector (detr/ deformable detr).