CenterTrack

CenterTrack copied to clipboard

Integrate CenterTrack with any detector

1

Hello, Thank you for this wonderful work and repo ! I would like to quickly understand how I can integrate this tracker with any other detector I have as I...

akshu281

I found that there is a file named `data_parallel.py` under the folder `src/lib/model/`. It really confused me because the class `DataParallel` defined there seems just a copy of `torch.nn.DataParallel`. So,...

anti-destiny

Train CenterNet with this code

1

Hi! It's a nice work!! I wonder if I can train a CenterNet for image detection on coco? I train the model with ''' python main.py ctdet --exp_id coco_dla_1x --batch_size...

zhangxgu

What does "if state_dict[k].shape[0] < state_dict[k].shape[0]:" mean

1

Hi, thanks for your great work! However, I'm confused during reading the code, especially this line: in /src/lib/model/model.py line 55: if state_dict[k].shape[0] < state_dict[k].shape[0]: I'm so confused about its purpose...

infZhou

Is testing results on MOT17 includes both Padestrian and Car?

1

Hi, could you please confirm if the final scores include the tracking for both pedestrians and cars on the MOT17 dataset? Thanks,

NeeluMadan

The below if statement in model.py (line 55) is always 'False'. **55: if state_dict[k].shape[0] < state_dict[k].shape[0]:** 56: model_state_dict[k][:state_dict[k].shape[0]] = state_dict[k] 57: else: 58: model_state_dict[k] = state_dict[k][:model_state_dict[k].shape[0]] Shouldn't one of the...

brjeon1

When the graphics card 1 is being used by others and the graphics card 2 is not being used, the video card 2 cannot be called for training, resulting in an overflow error

2

When there are 4 g of video memory left on the first video card and 11 g of video memory left on the second video card, you can train with...

lushbng

Evaluation for nuScenes dataset - Mismatched name length and levels in MultiIndex

1

Hi, I am trying to run the evaluation for nuScenes dataset > python test.py tracking,ddd --exp_id nuScenes_3Dtracking --load_model ../models/nuScenes_3Dtracking.pth --dataset nuscenes --track_thresh 0.1 --pre_hm and receive the following error: >...

jschroeder431

I want to debug mian.py , but when debugging to starting training... It can't be debugged any more, and there is no training progress bar, so it's stuck there all...

lushbng

Could anyone share a docker image or dockerfile that has can run this repo? After compile DCNv2 successful, still can not import DCN #53

3

After compile DCNv2 successful, still can not import DCN #53 My environment: Ubuntu 18.04, with Anaconda Python 3.6, CUDA 10.0.130, and PyTorch v1.0.0 Many thx!

sisrfeng