Disparity Estimation Using a Quad-Pixel Sensor

TL;DR

This paper introduces QPDNet, a novel neural network leveraging quad-pixel sensor data for improved depth estimation, outperforming existing stereo and dual-pixel methods.

Contribution

The paper presents a new disparity estimation network that exploits quad-pixel sensor information and a synthetic dataset generation pipeline.

Findings

QPDNet outperforms state-of-the-art stereo and DP methods.

Synthetic dataset effectively trains the disparity network.

Quad-pixel sensors provide rich information for depth estimation.

Abstract

A quad-pixel (QP) sensor is increasingly integrated into commercial mobile cameras. The QP sensor has a unit of 2$\times$2 four photodiodes under a single microlens, generating multi-directional phase shifting when out-focus blurs occur. Similar to a dual-pixel (DP) sensor, the phase shifting can be regarded as stereo disparity and utilized for depth estimation. Based on this, we propose a QP disparity estimation network (QPDNet), which exploits abundant QP information by fusing vertical and horizontal stereo-matching correlations for effective disparity estimation. We also present a synthetic pipeline to generate a training dataset from an existing RGB-Depth dataset. Experimental results demonstrate that our QPDNet outperforms state-of-the-art stereo and DP methods. Our code and synthetic dataset are available at https://github.com/Zhuofeng-Wu/QPDNet.