Object detection using reinforcement learning

Installation Dependencies:

• Python 2.7
• TensorFlow 1.2.0
• keras 2.0.5
• OpenCV-Python

Required Data

PASCAL VOC2007 and 2012 datasets Available at: http://host.robots.ox.ac.uk/pascal/VOC/index.html

Description

The RL agent learns to localise an object in an image of a given class by taking a series of actions from a discrete action space.

This RL implementation uses Q-learning with the Q-function being approximated with a deep fully connected network.

To be Implemented

• Detect multiple instances of an object in a given image (implement IoR mechanism)
• Classification score for detections

Further Ideas

• Lightweight classifier (Inception/Xception)
• Multiclass extension