Implementation of YOLO v3 object detector in Tensorflow (TF-Slim). Full tutorial can be found here.
Tested on Python 3.5, Tensorflow 1.11.0 on Ubuntu 16.04.
- YOLO v3 architecture
- Basic working demo
- Weights converter (util for exporting loaded COCO weights as TF checkpoint)
- Training pipeline
- More backends
To run demo type this in the command line:
- Download COCO class names file:
wget https://raw.githubusercontent.com/pjreddie/darknet/master/data/coco.names
- Download and convert model weights:
- Download binary file with weights from https://pjreddie.com/darknet/yolo/
- Run
python ./convert_weights.py
- Run
python ./demo.py --input_img <path-to-image> --output_img <name-of-output-image>
- Added weights converter from Darknet format to Tensorflow model checkpoint
- Updated demo.py script to use TF saved checkpoint
- Moved utility functions (load_weights, load_coco_names etc.) to utils.py
- Added data_format flag, possible options are: NCHW (works only on GPU) / NHWC (woks on both CPU and GPU)
- Merged PR36, which fixes the bug with bad stride in convolutional layer in YOLOv3-Tiny. Thank you @LucasMahieu.
- Merged PR with YOLOv3-Tiny model
- Bug fixes