Utilization of Deep Reinforcement Learning for saccadic-based object visual search

2016/10/20 by Tomasz Kornuta, Kamil Rocki, Kornuta, Tomasz +1
Computer Science · Engineering · #Industrial Vision Systems and Defect Detection #cs.CV #cs.LG

paper · pdf · doi:10.48550/arxiv.1610.06492

Paper submitted to special session on Machine Intelligence organized during 23rd International AUTOMATION Conference

arxiv created 2016/10/20 · arxiv updated 2016/10/21

Abstract

The paper focuses on the problem of learning saccades enabling visual object search. The developed system combines reinforcement learning with a neural network for learning to predict the possible outcomes of its actions. We validated the solution in three types of environment consisting of (pseudo)-randomly generated matrices of digits. The experimental verification is followed by the discussion regarding elements required by systems mimicking the fovea movement and possible further research directions.

Related

• Saccade Mechanisms for Image Classification, Object Detection and Tracking
• A Saccaded Visual Transformer for General Object Spotting
• An ocular biomechanics environment for reinforcement learning
• SAFER-AiD: Saccade-Assisted Foveal-peripheral vision Enhanced Reconstruction for Adversarial Defense
• Energy-Efficient Visual Search by Eye Movement and Low-Latency Spiking Neural Network