Efficient Depth Propagation in Videos with GPU-acceleration
By Manuel Ivancsics, Nicole Brosch, and Margrit Gelautz
Abstract
In this paper we propose an optimized semi-automatic approach for efficient 2D-to-3D video conversion. It is based on a conversion algorithm [1] that leverages segmentation and filtering techniques to propagate sparse depth information that was provided by a user. Our GPU acceleration of [1] significantly reduces the computation time of the original algorithm. Since the limited capacity of the GPU´s onboard memory hinders the parallel execution of large data such as videos, we additionally propose a temporally coherent clip-based 2D-to-3D conversion approach for long videos. Evaluations show that the proposed, optimized conversion approach is capable of generating high-quality results, while significantly reducing the execution time compared to the original, un-optimized approach.
Reference
M. Ivancsics, N. Brosch, M. Gelautz: "Efficient Depth Propagation in Videos with GPU-acceleration"; Poster: IEEE Visual Communications and Image Processing (IEEE VCIP) 2014, Malta; 12-07-2014; in: "IEEE Visual Communications and Image Processing (IEEE VCIP)", (2014), 4 pages.
Additional Information
- Corresponding master thesis: Effiziente Tiefenpropagierung in Videos mit GPU-Unterstützung
- Algorithm that is optimized: Segmentation-based depth propagation in videos
Downloads
Supplementary Material
| VCIP14: Supplementary Material for "Efficient Depth Propagation in Videos with GPU-acceleration" | 2.12 MB | PDF document | Download |
BibTeX
Click into the text area and press Ctrl+A/Ctrl+C or ⌘+A/⌘+C to copy the BibTeX into your clipboard… or download the BibTeX.