- Download ByteTrack Github Resources
- Amend the code
- Install requirements
- Justify two codes
- Downloas pretrain file of YOLOX
- Test
🔗 https://github.com/ifzhang/ByteTrack
Near 50 lines in set.py
Just add encoding='utf-8' behind "r" .
then enter the following in prompt
python setup.py develop
After finished, it would be like that.
pip install cython
pip install -e git+https://github.com/samson-wang/cython_bbox.git#egg=cython-bbox
pip install pycocotools
pip install -r requirements.txt
⚠ If you would run Bytetrack in GPU, plz do more step as following. ⚠
Visit the Official Pytorch to get the cmd to install pytorch of CUDA.
🔗 https://pytorch.org/get-started/locally/
nvidia-smi
PS. If you're sure about CUDA version, you can check it in this cmd.
📍 yolox/data/data_auguments.py
find preproc function to replace as the following.
def preproc(img, input_size, swap=(2, 0, 1)):
if len(img.shape) == 3:
padded_img = np.ones((input_size[0], input_size[1], 3), dtype=np.uint8) * 114
else:
padded_img = np.ones(input_size, dtype=np.uint8) * 114
r = min(input_size[0] / img.shape[0], input_size[1] / img.shape[1])
resized_img = cv2.resize(
img,
(int(img.shape[1] * r), int(img.shape[0] * r)),
interpolation=cv2.INTER_LINEAR,
).astype(np.uint8)
padded_img[: int(img.shape[0] * r), : int(img.shape[1] * r)] = resized_img
padded_img = padded_img.transpose(swap)
padded_img = np.ascontiguousarray(padded_img, dtype=np.float32)
return padded_img, r
📍 tools/demo_track.py
Delete "self.rgb_means, self.std"
🔗 https://github.com/Megvii-BaseDetection/YOLOX
You can choose any you prefer to download,then put it in root folder.
python tools/demo_track.py video -f exps/default/yolox_x.py -c yolox_x.pth --fp16 --fuse --save_result
If you successed, it would be like 👇👇
bandicam.2022-08-01.15-50-27-066.mp4
The output would be in "YOLOX_outputs/yolox_x/track_vis"
Here is the ouput I tested.
https://github.com/ifzhang/ByteTrack https://colab.research.google.com/drive/1bDilg4cmXFa8HCKHbsZ_p16p0vrhLyu0?usp=sharing#scrollTo=kcuzq_MJ1EmK https://blog.csdn.net/kuweicai/article/details/120873335