Renrui Zhang

@euminds Thanks for pointing out. We have fix this and release a new code base in a repo. Concerning 65.45% for Tip-Adapter, the released code would achieve 65.51% on my...

@heng-yin Thanks for your great patience! We have release the code for other datasets in the repo. Maybe you can follow their configs to implement on CIFAR100.

Thanks for your interest! I suppose if the query and values are within the same embedding space, e.g., both text features, they can directly calculate the affinity matching and produce...

That is a quite insightful question. I tried on some datasets with varying K for different categories. Generally, a larger K leads to higher classification accuracy for the corresponding category....

Thanks for your interest. 6 dimensions denote (x, y, l, t, t, b), representing the xy center and the distances of four box boundaries lrtb to the center.

The (l, r, t, b) is especially for monocular 3D object detection adopted by [MonoFlex](https://openaccess.thecvf.com/content/CVPR2021/papers/Zhang_Objects_Are_Different_Flexible_Monocular_3D_Object_Detection_CVPR_2021_paper.pdf), since the projected 3D center may not locate at the center of the 2D box....

@ysyf293 We refer to the tricks in GroupDETR to utilize multiple groups for queries, 11 by default, for more stable performance.

The following configs can achieve the best effect. We are still investigating how the two-stage/dab/dn/dino tricks can be used to improve MonoDETR. two_stage: False use_dab: False use_dn: False two_stage_dino: False...

The current code with configurations is only for KITTI val set. We will release the test code soon. Thanks for your waiting.

Thanks for your interest. The original code can reproduce the result in our paper. You can refer to the provided several logs and ckpts based on our original code in...