tensorflow-yolo-v3
tensorflow-yolo-v3 copied to clipboard
Implementation of YOLO v3 object detector in Tensorflow (TF-Slim)
tensorflow-yolo-v3
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.
Todo list:
- [x] YOLO v3 architecture
- [x] Basic working demo
- [x] Weights converter (util for exporting loaded COCO weights as TF checkpoint)
- [ ] Training pipeline
- [ ] More backends
How to run the demo:
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 desired weights:
- Full weights:
wget https://pjreddie.com/media/files/yolov3.weights
- Tiny weights:
wget https://pjreddie.com/media/files/yolov3-tiny.weights
- SPP weights:
wget https://pjreddie.com/media/files/yolov3-spp.weights
- Full weights:
- Run
python ./convert_weights.py
andpython ./convert_weights_pb.py
- Download binary file with desired weights:
- Run
python ./demo.py --input_img <path-to-image> --output_img <name-of-output-image> --frozen_model <path-to-frozen-model>
####Optional Flags
- convert_weights:
-
--class_names
- Path to the class names file
-
--weights_file
- Path to the desired weights file
-
--data_format
-
NCHW
(gpu only) orNHWC
-
-
--tiny
- Use yolov3-tiny
-
--spp
- Use yolov3-spp
-
--ckpt_file
- Output checkpoint file
-
- convert_weights_pb.py:
-
--class_names
1. Path to the class names file -
--weights_file
- Path to the desired weights file
-
--data_format
-
NCHW
(gpu only) orNHWC
-
-
--tiny
- Use yolov3-tiny
-
--spp
- Use yolov3-spp
-
--output_graph
- Location to write the output .pb graph to
-
- demo.py
-
--class_names
- Path to the class names file
-
--weights_file
- Path to the desired weights file
-
--data_format
-
NCHW
(gpu only) orNHWC
-
-
--ckpt_file
- Path to the checkpoint file
-
--frozen_model
- Path to the frozen model
-
--conf_threshold
- Desired confidence threshold
-
--iou_threshold
- Desired iou threshold
-
--gpu_memory_fraction
- Fraction of gpu memory to work with
-