吴强副教授学术交流
报告日期 Date: 2016年7月6日,星期三
报告时间 Time: 上午10:00
报告地点 Location: 福州大学物理与信息工程学院祥联厅
报告人: 吴强 (Qiang Wu), Ph.D. Associate Professor,University of Technology Sydney
报告内容:
Edges Inconsistency Evaluation Model for Color-guided Depth Map Enhancement
Abstract (摘要): Color-guided depth enhancement is to refine depth maps according to the assumption that the depth edge and the color edge at the corresponding location are consistent. Through all methods reported on color-guided depth enhancement, Markov Random Fields (MRF) including its variants is one of major approaches, which has dominated in this area for several years. However, the assumption above is not always true. The state-of-the-art solutions usually are to adjust the weighting coefficient inside the smoothness term of MRF model. Because there is no any method explicitly measuring the inconsistency occurring between the depth edge and the corresponding color edge, they cannot adaptively control the effect of guidance based on the color image when enhancing the depth map. Consequently, it causes various defects such as texture-copy artifacts and depth edges blurring. In this talk, quantitative measurement on such inconsistency is introduced. It is explicitly embed it into the weighting coefficient in the smoothness term. The proposed method demonstrates the promising results when compared with ten classic or state-of-the-art methods.