abawchen/tf-adnet-tracking

Deep Object Tracking Implementation in Tensorflow for 'Action-Decision Networks for Visual Tracking with Deep Reinforcement Learning(CVPR 2017)'

tf-adnet

Tensorflow Implementation of 'Action-Decision Networks for Visual Tracking with Deep Reinforcement Learning(CVPR 2017)'

GIF	Description
	OTB100 Dataset Freeman1.

• Green : Ground Truth, Blue : ADNet-Fast Tracking Result

Implementations

• Network Architecture

• Training Code

  • Reproduce Paper Result

• Inference Code

  • Converting Original Weights to Tensorflow

  • Online Learning(finetuning)

  • Faster Inference

Run

OTB100 Dataset

Create a virtualenv in which we can install the dependencies (recommanded):

$ virtualenv env
$ source env/bin/activate

Install dependencies:

$ pip install -r requirements.txt

$ python runner.py by_dataset  --vid-path=./data/freeman1/

References

• (Korean, 한국어) Review : http://openresearch.ai/t/adnet-action-decision-networks-for-visual-tracking-with-deep-reinforcement-learning/123
• Action-Decision Networks for Visual Tracking with Deep Reinforcement Learning (CVPR201) : http://openaccess.thecvf.com/content_cvpr_2017/papers/Yun_Action-Decision_Networks_for_CVPR_2017_paper.pdf
• ADNet Implmentation in Matlab : https://github.com/hellbell/ADNet/