SymmNet: A Symmetric Convolutional Neural Network for Occlusion Detection

TL;DR

SymmNet is a novel symmetric CNN that directly detects stereo and motion occlusions from image pairs without prior disparity or optical flow estimation, achieving state-of-the-art results.

Contribution

It introduces a symmetric neural network architecture that jointly detects occlusions from stereo images or videos, bypassing traditional disparity or flow computation.

Findings

Achieves state-of-the-art occlusion detection accuracy

Effectively detects both stereo and motion occlusions

Joint learning improves detection performance

Abstract

Detecting the occlusion from stereo images or video frames is important to many computer vision applications. Previous efforts focus on bundling it with the computation of disparity or optical flow, leading to a chicken-and-egg problem. In this paper, we leverage convolutional neural network to liberate the occlusion detection task from the interleaved, traditional calculation framework. We propose a Symmetric Network (SymmNet) to directly exploit information from an image pair, without estimating disparity or motion in advance. The proposed network is structurally left-right symmetric to learn the binocular occlusion simultaneously, aimed at jointly improving both results. The comprehensive experiments show that our model achieves state-of-the-art results on detecting the stereo and motion occlusion.