Skip to content

Fine-tuning ByteTrack for a Study at the School of Cognitive Science, IPM.

License

Notifications You must be signed in to change notification settings

aryan-f/ByteTrack

 
 

Repository files navigation

ByteTrack for Caltech-UCSD Birds

Tracking performance

Results

Not yet reported.

Installation

1. Install the main dependencies.

git clone https://github.com/ifzhang/ByteTrack.git
cd ByteTrack
pip3 install -r requirements.txt
python3 setup.py develop

2. Install pycocotools.

pip3 install cython; pip3 install 'git+https://github.com/cocodataset/cocoapi.git#subdirectory=PythonAPI'

3. Others

pip3 install cython_bbox

Data preparation

Download Caltech-UCSD Birds and save it in the datasets directory. The structure of the folder must look like this:

datasets
   |——————CaltechBirds
   |        └——————bounding_boxes.txt
   |        └——————classes.txt
   |        └——————image_class_labels.txt
   |        └——————images.txt
   |        └——————train_test_split.txt
   |        └——————images/

Then, you need to turn the dataset to COCO format:

cd <ByteTrack_HOME>
python tools/convert_birds_to_coco.py

About

Fine-tuning ByteTrack for a Study at the School of Cognitive Science, IPM.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 84.0%
  • C++ 14.9%
  • Other 1.1%