Abstract
Light field imaging based on microlens arrays - a.k.a. holoscopic, plenoptic, and integral imaging - has currently risen up as a feasible and prospective technology for future image and video applications. However, deploying actual light field applications will require identifying more powerful representation and coding solutions that support emerging manipulation and interaction functionalities. In this context, this paper proposes a novel scalable coding approach that supports a new type of scalability, referred to as Field of View (FOV) scalability, in which enhancement layers can correspond to regions of interest (ROI). The proposed scalable coding approach comprises a base layer compliant with the High Efficiency Video Coding (HEVC) standard, complemented by one or more enhancement layers that progressively allow richer versions of the same light field content in terms of content manipulation and interaction possibilities, for the whole scene or just for a given ROI. Experimental results show the advantages of the proposed scalable coding approach with ROI support to cater for users with different preferences/requirements in terms of interaction functionalities.
Highlights
In the context of Light Field (LF) imaging, the approach based on a single-tier camera with a Microlens Array (MLA) [2] has become a promising approach with application in many different areas, such as 3D television [3], richer photography capturing [4], biometric recognition [5], and medical imaging [6]
The LF data is firstly organized into several layers, in which higher layers typically correspond to regions of the LF image with wider Field of View (FOV)
Each enhancement layer is encoded with the proposed LF enhancement layer codec seen in Fig. 4, which is based on the High Efficiency Video Coding (HEVC) architecture and may explore spatial and inter-layer redundancy through self-similarity (SS) [11] and inter-layer predictions; a detailed description of the modules used for inter-layer prediction is provided in [13]
Summary
Recent advances in the manufacturing of optics and imaging sensors made it possible to have richer forms of visual data, where spatial information about three-dimensional (3D) scenes is represented in addition to angular viewing direction — the four-dimensional (4D) light field/radiance sampling [1]. The challenge to provide a LF representation with convenient spatial resolution and viewing angles requires handling a huge amount of data and, efficient coding becomes of utmost importance Another key requirement when designing an efficient LF representation and coding solution is to facilitate future interactive LF media applications supporting new manipulation functionalities. With the LF content can promptly start visualizing and manipulating the LF content, by extracting only the adequate bitstream subsets (which fit in the available bitrate) To better handle this type of application scenarios, the coding architecture proposed in [13] is here extended to enable easy support for Region of Interest (ROI) coding [14] in the enhancement layers.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
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 callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
