Real-time 2D to 3D video conversion

Abstract

We present a real-time implementation of 2D to 3D video conversion using compressed video. In our method, compressed 2D video is analyzed by extracting motion vectors. Using the motion vector maps, depth maps are built for each frame and the frames are segmented to provide object-wise depth ordering. These data are then used to synthesize stereo pairs. 3D video synthesized in this fashion can be viewed using any stereoscopic display. In our implementation, anaglyph projection was selected as the 3D visualization method, because it is mostly suited to standard displays.

Real-Time 2D-to-3D Conversion for 3DTV using Time-Coherent Depth-Map Generation Method

“…Real-time conversion algorithms have been proposed by using multi-threaded CPUs or GPUs for parallel processing [12], [16]. The motion vectors in the MPEG 4 standard are also used to generate real-time depth maps [17]. Moreover, depth flickering should be reduced to mitigate eye fatigue during extended watching sessions.…”

A competitive study of the pseudoflow algorithm for the minimum s–t cut problem in vision applications

“…V enus T sukuba 2) Multi-view reconstruction: A 3D reconstruction is a fundamental problem in computer vision with a significant number of applications (for recent examples see [6], [7], [37]). Specifically, graph theory based algorithms for this problem were reported in [38]- [40].The input for the multi-view reconstruction problem is a set of 2D images of the same scene taken from different perspectives.…”

Automatic 2D-to-3D video conversion technique based on depth-from-motion and color segmentation

“…Thus, the main purpose of the 2D-to-3D stereoscopic video conversion system is to generate additional views from monoscopic video input. The basic structure of the proposed automatic 2D-to-3D video conversion system using block-matching based depth from motion estimation [7] and color based region segmentation is shown in Fig. 1.…”

“…2(a), in which grey level 0 indicates that furthest distance from camera and the grey level 255 specifying the nearest distance. To achieve good depth map quality in the proposed system, the depth map of each frame is first estimated by block-matching based motion estimation [7] and then fused with color based region segmented image. The basic principle is underlying on the motion parallax, near objects move faster than far objects, and thus relative motion can be used to estimate the depth map.…”

“…Basically, generation of 3D video from monoscopic 2D video input source [2][3][4][5][6][7][8][9][10] have been investigated for many years. Most of them are based on an estimated depth map of each frame and then using DIBR (Depth Image Based Rendering) [12] to synthesized the additional views.…”