Abstract

While eight-bit playback and display devices will be dominating the marketplaces in the near future, superior visual quality by high bit-depth videos is desirable for applications such as high standard entertainment and healthcare. Hence, conventional eight-bit and high bit-depth digital imaging systems will coexist in marketplaces. Content distributors supporting both of them need to provide different contents for different users, e.g., simulcastly code the different representations for the same video content. This requires more storage or bandwidth for video content delivery. Bit-depth scalability is an efficient tool to solve this problem. However, video coding techniques can allow flexible usage of various versions of the same visual content that may have spatial resolutions and even alterations in color. In this paper, we propose a bit-depth scalable coding solution that is compatible to the scalable extension of H.264/AVC, also known as Scalable Video Coding (SVC). The proposed bit-depth scalable coding is capable of providing an 8-bit AVC main profile or high profile compliant base layer which is multiplexed with a high bit depth (e.g., 10-, 12-, or up to 14-bit) enhancement layer through macroblock level inter-layer bit-depth prediction. New decoding processes for inter-layer bit-depth prediction are introduced to enable bit-depth scalability. Combination with other types of scalability: temporal, spatial and SNR scalability, as well as single-loop decoding is also supported since our algorithm is implemented based on the up-to-date SVC reference software. Furthermore, the proposed solution supports adaptive inter-layer prediction to determine whether or not the inter-layer bit-depth prediction shall be invoked for each Macroblock. The coding efficiency of the proposed bit-depth scalable coding can be further improved by incorporating advanced inter-layer bit-depth prediction algorithms. Experimental results are presented for 8-bit to high bit (10 or 12) scalability and also the combined bit-depth and spatial scalability.

Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call