DAMO-YOLO: a fast and accurate object detection method with some new techs, including NAS backbones, efficient RepGFPN, ZeroHead, AlignedOTA, and distillation enhancement.
[X] I have pulled the latest code of main branch to run again and the bug still existed. 我已经拉取了主分支上最新的代码，重新运行之后，问题仍不能解决。
[X] I have read the README carefully and no error occured during the installation process. (Otherwise, we recommand that you can ask a question using the Question template) 我已经仔细阅读了README上的操作指引，并且在安装过程中没有错误发生。（否则，我们建议您使用Question模板向我们进行提问）
Search before reporting
[X] I have searched the DAMO-YOLO issues and found no similar bugs. 我已经在issue列表中搜索但是没有发现类似的bug报告。
OS
Ubuntu
Device
Nvidia T4
CUDA version
12.2
TensorRT version
8.6.1.6
Python version
3.10
PyTorch version
2.0.1+cu117
torchvision version
2.0.1+cu117
Describe the bug
tensorRT model for tensorRT 8 outputs incorrect bounding boxes. Technically, since it's incorrectly interpreting the input tensor, it could be wrong out the rest of the output too.
Before Reporting
[X] I have pulled the latest code of main branch to run again and the bug still existed. 我已经拉取了主分支上最新的代码，重新运行之后，问题仍不能解决。
[X] I have read the README carefully and no error occured during the installation process. (Otherwise, we recommand that you can ask a question using the Question template) 我已经仔细阅读了README上的操作指引，并且在安装过程中没有错误发生。（否则，我们建议您使用Question模板向我们进行提问）
Search before reporting
OS
Ubuntu
Device
Nvidia T4
CUDA version
12.2
TensorRT version
8.6.1.6
Python version
3.10
PyTorch version
2.0.1+cu117
torchvision version
2.0.1+cu117
Describe the bug
tensorRT model for tensorRT 8 outputs incorrect bounding boxes. Technically, since it's incorrectly interpreting the input tensor, it could be wrong out the rest of the output too.
python tools/converter.py -f configs/damoyolo_tinynasL45_L.py -c model.pth --batch_size 1 --img_size 1024 --trt --end2endThis was discussed in #102
It is fixed in PR #113 which changes box_coding from 1 (BoxCenterSize) to 0 (BoxCorner). See https://github.com/NVIDIA/TensorRT/tree/release/8.6/plugin/efficientNMSPlugin for more information on the parameters.
To Reproduce
Run
python tools/converter.py -f configs/damoyolo_tinynasL45_L.py -c best.pth --batch_size 1 --img_size 1024 --trt --end2end --trt_evalThe evaluation is 0%.
I also use demo command to predict some images. The output show that the bounding box seem randomly.
Hyper-parameters/Configs
No response
Logs
No response
Screenshots
No response
Additional
No response