Object Detection

This is the implementation of YOLOv3 for object detection in Tensorflow. It contains complete code for preprocessing, training and test. Besides, this repository is easy-to-use and can be developed on Linux and Windows.

YOLOv3 : Redmon, Joseph, and Ali Farhadi. "Yolov3: An incremental improvement." arXiv preprint arXiv:1804.02767 (2018).

Getting Started

1 Prerequisites

• Python 3.x
• Tensorflow 1.x
• Opencv-python
• Pandas

2 Define your class names

Download and unzip this repository.
cd ../YOLOv3/label
Open the label.txt and revise its class names as yours.

3 Prepare images

Copy your images and annotation files to directories ../YOLOv3/data/annotation/images and ../YOLOv3/data/annotation/images/xml respectively, where the annotations should be obtained by a graphical image annotation tool and saved as XML files in PASCAL VOC format.
cd ../YOLOv3/code
run
python spilt.py
Then train and val images will be generated in ../YOLOv3/data/annotation/train and /YOLOv3/data/annotation/test directories, respectively.

4 Anchor clusters using K-means

Run K-means clustering on the training set bounding boxes to automatically find good anchors.
cd ../YOLOv3/code
run
python anchor_cluster.py
Anchors generated by K-means are saved in the directory ../YOLOv3/anchor/anchor.txt. Belows are same outputs when running K-means:

Iter = 1/20, Average IoU = 0.719983, is current optimal anchors.
Iter = 2/20, Average IoU = 0.733096, is current optimal anchors.
Iter = 3/20, Average IoU = 0.73589, is current optimal anchors.
Iter = 4/20, Average IoU = 0.736503, is current optimal anchors.
Iter = 5/20, Average IoU = 0.736472
Iter = 6/20, Average IoU = 0.736157
Iter = 7/20, Average IoU = 0.735872
Iter = 8/20, Average IoU = 0.735478
...................................
...................................
...................................
Iter = 19/20, Average IoU = 0.732432
Iter = 20/20, Average IoU = 0.73226

5 Train model using Tensorflow

The model parameters, training parameters and eval parameters are all defined by parameters.py.
cd ../YOLOv3/code
run
python train.py
The model will be saved in directory ../YOLOv3/model/checkpoint, and some detection results are saved in ../YOLOv3/pic.

6 Visualize model using Tensorboard

cd ../YOLOv3
run
tensorboard --logdir=model/
Open the URL in browser to visualize model. Below is the graph of my model:

Examples

Here are some detection examples in my dataset: