NonMaxSuppression-with-Reinforcement-Learning

A Reinforcement Learning Agent Doing Non Maximum Suppression For Object detection.

Setup

Clone the repository:

$ git clone https://github.com/blurry-mood/NonMaxSuppression-with-Reinforcement-Learning

Install Requirements (or first create a virual environment):

$ cd NonMaxSuppression-with-Reinforcement-Learning
$ pip3 install -r requirements

Dataset

The used dataset is WIDER FACE. It can be found here.
Please download the training, validation, testing images and the face annotations file, and place them inside dataset/.
Finally execute this command:

$ cd NonMaxSuppression-with-Reinforcement-Learning/dataset
$ sh setup.sh

Train

Training the agent is done by executing this command:

$ cd NonMaxSuppression-with-Reinforcement-Learning/src
$ python3 train_agent.py

After every episode, the agent will be saved the model to artifacts/dqn.pth. A UI will be shown to visualize the training process. At every step in the episode, chosen bounding-boxes will be visualized.

Citation

@inproceedings{yang2016wider,
                Author = {Yang, Shuo and Luo, Ping and Loy, Chen Change and Tang, Xiaoou},
                Booktitle = {IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
                Title = {WIDER FACE: A Face Detection Benchmark},
                Year = {2016}}

@inproceedings{deng2019retinaface,
                title={RetinaFace: Single-stage Dense Face Localisation in the Wild},
                author={Deng, Jiankang and Guo, Jia and Yuxiang, Zhou and Jinke Yu and Irene Kotsia and Zafeiriou, Stefanos},
                booktitle={arxiv},
                year={2019}