Low-complexity Depth Map Generation for Real-time 2D-to-3D Video Conversion

A low-complexity depth generation algorithm for 3D conversion of 2D videos is presented. For temporal consistency in global depth, a pattern-based depth generation method is introduced. In addition, refinement of depth from the combined depth cues of texture and motion is also addressed. Experimental results show that the proposed method outperforms previous algorithms in terms of the complexity and subjective quality.