Research On Coding And Compression Algorithm Of 3D Video Depth Data

3D data is widely used in computer-aided design,entertainment,virtual reality,industrial rapid prototyping and 3D printing.The improvement of computer computing speed and the continuous development of optical imaging technology have created conditions for the real-time acquisition of 3D data.At the same time,it also puts forward new requirements for compressed storage and transmission technologies in the later stage of 3D data.On the basis of in-depth study of various 3D data encoding and compression algorithms,this paper improves the algorithm of virtual structured 3D data encoding algorithm,and improve the alternative CVBF algorithm on scalable regions.The specific research contents are as follows:1.Virtual structured light 3D data encoding algorithm In the process of data encoding and decoding,the fringes storing phase-level sub-indexes may cause phase error due to quantization errors at the edges,eventually cause decoding depth information errors.To solve this problem,this paper proposes a relative phase unwrapping assisted absolute phase unwrapping method to suppress deep decoding errors caused by quantization errors.In the process of decoding the fringe's truncated phase into a continuous phase to complete the depth information decoding,the relative phase unwrapping and absolute phase unwrapping are used to complete the truncated phase unwrapping simultaneously.According to the characteristics of the fixed phase difference value obtained by the two methods,the storage phase is fixed.The quantization error at the fringe edge of the level is corrected.The experimental results show that this method can effectively improve the data fidelity of the virtual structured light 3D data encoding algorithm,and the average root mean square error of data decoding is reduced by 10.8 times.2.When HEVC is used to compress the depth sequence of 3D video,it will cause artifacts on the edges of the depth map,which will eventually lead to a void in the synthesized new viewpoint,affect the video viewing quality.CFBV algorithm uses alternative pixels instead of depth edge untrusted pixels,which is a more effective and simple depth edge filtering algorithm.In order to obtain more accurate depth data,this paper proposes an alternative pixel extraction algorithm based on scalable region to improve the robustness of CVBF algorithm.In the filtering process,the values of some candidate pixels are obtained by averaging the areas with high credibility.Experimental results show that using the improved filtering method,the PSNR of the synthetic virtual viewpoint is 0.08 d B higher than that of the CVBF algorithm.The filtering method can effectively improve the reliability of compression-encoded depth sequence boundary data,thereby improving the quality of virtual viewpoint synthesis.