mmtracking

mmtracking copied to clipboard

I am training my custom dataset on [video object detection](https://github.com/open-mmlab/mmtracking/tree/master/configs/vid/temporal_roi_align). My dataset was in VOC format. I have written a custom [script](https://github.com/danial880/mmtrack-custom/blob/main/voc_to_cocovid_resized.py) to convert VOC to COCOVid format. I have...

danial880

Maybe a small bug about test progress bar in multi_gpu_test().

1

During distributed training, sot model use multi_gpu_test() func to test all videos on several GPUs. But the progress bar is updated on rank 0. https://github.com/open-mmlab/mmtracking/blob/e79491ec8f0b8c86fda947fbaaa824c66ab2a991/mmtrack/apis/test.py#L160C34-L160C34 The number of frames on...

MisakaMikt

AssertionError: The number of person ids in the training set must be greater than the number of person ids in the sample.

2

Thanks for your error report and we appreciate it a lot. **Checklist** 1. I have searched related issues but cannot get the expected help. 2. The bug has not been...

GerogeLee0115

Which MMCV version should I use?

4

I try to install mmtracking with cuda11.7 and PyTorch 1.13.1. First, I installed mmcv=2.0.0, and run the demo `python demo/demo_mot_vis.py configs/mot/deepsort/sort_faster-rcnn_fpn_4e_mot17-private.py --input demo/demo.mp4 --output mot.mp4` `File "/home/xxx/code/mmtracking/mmtrack/__init__.py", line 71` it...

Kev1nZheng

AssertionError: MMCV==2.0.1 is used but incompatible. Please install mmcv>=1.3.17, <2.0.0.

3

After installing the compatible version , it suggesting . Even though getting the error.

tericalpha

Dear community, I am currently using mmpose with mmcv 2.1.0. but I also want to run mmTracking but I got an error related to mmcv version, mmTrack require mmcv>=1.3.17,

DiegoHerrera1890

Hi, I have a a question regarding MixFormer. I'm wondering if it's possible and how easy it would be to modify the memory to include two static targets instead of...

ekazakos

When training the ByteTrack using BDD100K dataset, detection accuracy downs to nearly zero after adding L1 Loss of YOLOX

4

I train the ByteTrack using official config, only modifying some hyperparameters and the input dataset. And it performs well before adding the L1 Loss.  After adding the L1 Loss,...

zengwz

Can mmtracking now adapt to the BDD100K dataset? How do we adapt?

yinxingshu41

Use pretrained RTMDet as detector

1

I'm trying to use `RTMDet` as the detector and use SORT to track a video with `demo_vid.py`, but I get the following errors: ``` 03/07 23:32:17 - mmengine - WARNING...

tctco