Depth Map Estimation Using Census Transform for Light Field Cameras

Abstract

SUMMARYDepth estimation for a lense-array type light field camera is a challenging problem because of the sensor noise and the radiometric distortion which is a global brightness change among sub-aperture images caused by a vignetting effect of the micro-lenses. We propose a depth map estimation method which has robustness against sensor noise and radiometric distortion. Our method first binarizes sub-aperture images by applying the census transform. Next, the binarized images are matched by computing the majority operations between corresponding bits and summing up the Hamming distance. An initial depth obtained by matching has ambiguity caused by extremely short baselines among sub-aperture images. After an initial depth estimation process, we refine the result with following refinement steps. Our refinement steps first approximate the initial depth as a set of depth planes. Next, we optimize the result of plane fitting with an edge-preserving smoothness term. Experiments show that our method outperforms the conventional methods.

Rapid Light Field Depth Estimation with Semi-Global Matching

“…it should not be noisy). Without boundaries, Census-based configuration has somewhat similar to the method of Tomioka et al [43] in terms of aggregation principles.…”

Fast Depth Estimation for Light Field Cameras

“…Multi-label optimization was then performed using graph cuts to enhance the initial estimation. Tomioka et al [8] calculated the matching cost based on census transform to withstand degradation caused by sensor noise and vignetting. Their refinement step minimizes an objective function that consists of a data term and an edge-preserving smoothness term.…”

Learning Multi-modal Information for Robust Light Field Depth Estimation

“…Light field depth estimation has become a popular research topic and plays a more and more important role in a wide range of applications, such as scene reconstruction [18], image super-resolution [31,41], object tracking [42], saliency detection [30], image segmentation [47]. According to the input data type, existing depth estimation methods from light field can be categorized into three types: depth estimation based on epipolar plane images (EPIs) [19,12], depth estimation based on sub-aperture images (sub-apertures) [37,40] and depth estimation based on the focal stack [28,36,23]. The EPIs-based depth estimation explores geometry structures in EPIs to capture the depth information.…”