search

megvii-research / PETR

[ECCV2022] PETR: Position Embedding Transformation for Multi-View 3D Object Detection & [ICCV2023] PETRv2: A Unified Framework for 3D Perception from Multi-Camera Images
Other
869 stars 132 forks source link

> > > Hi, #142

Open Vendulamrdka95 opened 1 year ago

Vendulamrdka95 commented 1 year ago 
          > > > Hi,

(1) When use mmdet1.0, have you notice here #71 (comment) . The reverse_angle must be False in GlobalRotScaleTransImage. (2) Yes, when set with_position=False, it's a result in ablation study. image When set with_position=False, the intrinsics and extrinsics are not used in model. In fact, PETR can work without intrinsics and extrinsics, benefiting from global attention. The low performance is mainly due to ResizeCropFlipImage and GlobalRotScaleTransImage. These data augmentation greatly change the intrinsics and extrinsics during the training process, and the network can't overfit the parameters of the data set. Once these augmentations are removed, resnet50 should obtain the peformance more than 27% mAP. But we don't think it's meaningful to over-fit the dataset.

I have noticed StreamPETR still set reverse_angle=True but they use mmdet3d=1.0.0rc6, have I missed something?

The rotate matrix is different.

Thanks, got it. 👍

Originally posted by @xiaosu-zhu in https://github.com/megvii-research/PETR/issues/86#issuecomment-1724719259

Vendulamrdka95 commented 1 year ago 
          > > > Hi,

(1) When use mmdet1.0, have you notice here #71 (comment) . The reverse_angle must be False in GlobalRotScaleTransImage. (2) Yes, when set with_position=False, it's a result in ablation study. image When set with_position=False, the intrinsics and extrinsics are not used in model. In fact, PETR can work without intrinsics and extrinsics, benefiting from global attention. The low performance is mainly due to ResizeCropFlipImage and GlobalRotScaleTransImage. These data augmentation greatly change the intrinsics and extrinsics during the training process, and the network can't overfit the parameters of the data set. Once these augmentations are removed, resnet50 should obtain the peformance more than 27% mAP. But we don't think it's meaningful to over-fit the dataset.

I have noticed StreamPETR still set reverse_angle=True but they use mmdet3d=1.0.0rc6, have I missed something?

The rotate matrix is different.

Thanks, got it. 👍

Originally posted by @xiaosu-zhu in #86 (comment)

          > > > Hi,

(1) When use mmdet1.0, have you notice here #71 (comment) . The reverse_angle must be False in GlobalRotScaleTransImage. (2) Yes, when set with_position=False, it's a result in ablation study. image When set with_position=False, the intrinsics and extrinsics are not used in model. In fact, PETR can work without intrinsics and extrinsics, benefiting from global attention. The low performance is mainly due to ResizeCropFlipImage and GlobalRotScaleTransImage. These data augmentation greatly change the intrinsics and extrinsics during the training process, and the network can't overfit the parameters of the data set. Once these augmentations are removed, resnet50 should obtain the peformance more than 27% mAP. But we don't think it's meaningful to over-fit the dataset.

I have noticed StreamPETR still set reverse_angle=True but they use mmdet3d=1.0.0rc6, have I missed something?

The rotate matrix is different.

Thanks, got it. 👍

Originally posted by @xiaosu-zhu in #86 (comment)