Adaptive 3D rendering based on region-of-interest

Abstract

ABSTRACT3D processing techniques are really promising. However, several hurdles have to be overcome. In this paper, two of them are examined. The first is related to the high disparity management. It is currently not well mastered and its impact is strong for viewing 3D scene on stereoscopic screens. The second concerns the salient regions of the scene. These areas are commonly called Region-Of-Interest (RoI) in the image processing domain. The problem appears when there are more than one region-of-interest in a video scene. Indeed, it is then complicated for the eyes to scan them and especially if the depth difference between them is high. In this contribution, the 3D experience is improved by applying some effects related to RoIs. The shift between the two views is adaptively adjusted in order to have a null disparity on a given area in the scene. In the proposed approach, these areas are the visually interesting areas. A constant disparity on the salient areas improves the viewing experience over the video sequence.

Computational Model of Stereoscopic 3D Visual Saliency

“…In a stereoscopic 3D display system, depth information is usually represented by means of a disparity map which shows the parallax of each pixel between the left-view image and the right-view image. In the literature, a disparity map is usually directly adopted as depth information [31]. However, we propose too add a transformation from a disparity map to a depth map, which represents perceived depth in unit of length, in the chain of 3D visual attention modeling, since even the same disparity value corresponds to different perceived depth depending on the viewing condition.…”

“…To overcome these problems, we adjusted the depth range of the scene by using the method proposed by Chamaret et al [31]. We shifted the left view to left, the right view to the right.…”

“…This type of models (e.g. [29], [30] and [31]) does not contain any depth-map-based feature-extraction processes. Apart from detecting the salient areas by using 2D visual features, these models share a same step in which depth information is used as the weighting factor of the 2D saliency.…”

Top-down Spatial Attention for Visual Search: Novelty Detection-Tracking Using Spatial Memory with a Mobile Robot

“…As stated by Wang et al [32], most of the 3D based models utilizes depth image from three perspectives [32]: i) depth weighting [33], ii) depth saliency [34], and iii) stereo vision models that compute and use disparity [35]. All the studies stated until now, they were generally space based approaches by comparing one position to another based on 2D color images or depth images for 3D models.…”

“…On the other hand, several approaches combined the 3D data integration to color based models for saliency computation [32][33][34][35][36][37]. As stated by Wang et al [32], most of the 3D based models utilizes depth image from three perspectives [32]: i) depth weighting [33], ii) depth saliency [34], and iii) stereo vision models that compute and use disparity [35].…”