Robust depth enhancement based on texture and depth consistency

Abstract

With advances in three-dimension television (3DTV) technology, accurate depth information for 3DTV broadcasting has gained much attention recently. The depth map, either retrieved by stereo matching or captured by the RGB-D camera, is mostly with lower resolution and often with noisy or missing values than the texture frame. How to effectively utilise highresolution texture image to enhance the corresponding depth map becomes an important and inevitable approach. In this study, the authors propose texture similarity-based hole filling, texture similarity-based depth enhancement and rotating counsel depth refinement to enhance the depth map. Thus, the proposed depth enhancement system could suppress the noise, fill the holes and sharpen the object edges simultaneously. Experimental results demonstrate that the proposed system provides a superior performance, especially around the object boundary comparing to the state-of-the-art depth enhancement methods.

Depth map artefacts reduction: a review

“…This filter separates the depth map into low and high frequencies applying inadequately the texture synthesis for low frequencies and inpainting technique on high frequencies. In [44], an algorithm is proposed for depth enhancement. Texture similarity approach is adopted for inpainting and denoising image.…”

Adaptive colour‐guided non‐local means algorithm for compound noise reduction of depth maps

“…Hui et al [34] proposed a method based on the RGB‐D guided filter for filtering raw depth maps and filling the missing regions. In [35], the authors proposed a depth enhancement method in which hole filling and denoising of the depth map are performed on the basis of a texture similarity approach. However, these methods are not sufficiently robust in terms of depth sharpness.…”

Texture Synthesis Repair of RealSense D435i Depth Images with Object-Oriented RGB Image Segmentation

“…However, a mosaic effect easily appeared in the repaired area. Chang et al [ 39 ] proposed hole filling based on texture similarity, depth enhancement based on texture similarity, and depth refinement by a rotation consultant to enhance the depth map. The proposed depth enhancement system can simultaneously suppress noise, fill holes, and sharpen target edges.…”