Quantized Census for Stereoscopic Image Matching

Abstract

Abstract-Current depth capturing devices show serious drawbacks in certain applications, for example ego-centric depth recovery: they are cumbersome, have a high power requirement, and do not portray high resolution at near distance. Stereo-matching techniques are a suitable alternative, but whilst the idea behind these techniques is simple it is well known that recovery of an accurate disparity map by stereo-matching requires overcoming three main problems: occluded regions causing absence of corresponding pixels; existence of noise in the image capturing sensor and inconsistent color and brightness in the captured images.We propose a modified version of the Census-Hamming cost function which allows more robust matching with an emphasis on improving performance under radiometric variations of the input images.

Data‐driven recovery of hand depth using CRRF on stereo images

“…Matching‐cost features , : the implementation used five matching‐cost functions: sum of absolute difference (SAD), sum of squared differences (SSD), normalised cross‐correlation (NCC), quantised census (QC) and zero‐mean sum of ADs (ZSADs). The reader is referred to [28] for details on these cost functions. These cost measures were chosen because of their prominence, computation cost and simplicity.…”

Conditional Regressive Random Forest Stereo-Based Hand Depth Recovery

“…The reader is referred to [1] for details on these cost functions. These cost measures were chosen because of their prominence, computation cost and simplicity.…”

HandyDepth: Example-based stereoscopic hand depth estimation using Eigen Leaf Node Features

“…The first stage in our framework involves building a database of disparity and depth images for different hand poses to train the regressive framework. Disparity is estimated from a stereo image pair using a robust stereo matching cost function (Quantized Census) [1]. Simultaneously, a depth image is acquired using a RGBD camera.…”