yingfei liu

Hi， The normalization range ( [-61.2, -61.2, -10.0, 61.2, 61.2, 10.0] ) don't need to be consistent with detection range ( [-51.2, -51.2, -5.0, 51.2, 51.2, 3.0] ) . Normalization...

Hi， You can refer to https://github.com/megvii-research/PETR/issues/75#issuecomment-1323666167. The result on val set with "dd3d_det_final.pth" may slight higher than the test result, which is overfitting and abnormal. "fcos3d_vovnet_imgbackbone-remapped.pth" is trained with FCOS3D...

> @yingfei1016 Thanks for your kind reply! I am clear about the current pre-training settings. By the way, to evaluate on the nuScenes **val** set, have you ever tried this...

"2f" is only use to debug since "30f" loads very slowly. It can sample from the "30f". ``` # ------------------------------------------------------------------------ # Copyright (c) 2022 megvii-model. All Rights Reserved. # ------------------------------------------------------------------------...

Hi， Can you provide the test pipeline and augmentation parameters (https://github.com/megvii-research/PETR/blob/main/projects/configs/petrv2/petrv2_vovnet_gridmask_p4_800x320.py#L150).

Hi， The "resize_lim" must be scaled with "final_dim". When use (640, 1600), the "resize_lim" should be (0.94, 1.25).

Hi， You are right, "resize_lim" is not used during inference. I re-checked the test pipeline and found no problem. By default, the test pipeline does not read gt boxes. In...

This will be open source later, but related colleagues are on vacation this week.