Light Field (LF) imaging empowers many attractive applications by simultaneously recording spatial and angular information of light rays. In order to meet the challenges of LF storage and transmission, many view reconstruction-based LF compression methods are put forward. However, occlusion issue and under-exploitation of LF rich structure information limit the view reconstruction qualities, which further influence LF compression efficiency. In order to alleviate these problems, in this paper, we propose a geometry-aware view reconstruction network for LF compression. In our method, only sparsely-sampled LF views are encoded, which are further used as priors to reconstruct the un-sampled LF views at the decoder side. The proposed reconstruction process contains two stages including geometry-aware reconstruction and texture refinement. The geometry-aware reconstruction stage utilizes a multi-stream framework, which can fully explore LF spatial-angular, location and geometry information. The texture refinement stage can adequately fuse such rich LF information to further improve LF reconstruction quality. Comprehensive experimental results validate the superiority of the proposed method. The rate-distortion performance and the perceptual quality of reconstructed views further demonstrate that the proposed method can save more bitrate while increasing LF reconstruction quality.

Light Field (LF) imaging empowers many attractive applications by simultaneously recording spatial and angular information of light rays. In order to meet the challenges of LF storage and transmission, many view reconstruction-based LF compression methods are put forward. However, occlusion issue and under-exploitation of LF rich structure information limit the view reconstruction qualities, which further influence LF compression efficiency. In order to alleviate these problems, in this paper, we propose a geometry-aware view reconstruction network for LF compression. In our method, only sparsely-sampled LF views are encoded, which are further used as priors to reconstruct the un-sampled LF views at the decoder side. The proposed reconstruction process contains two stages including geometry-aware reconstruction and texture refinement. The geometry-aware reconstruction stage utilizes a multi-stream framework, which can fully explore LF spatial-angular, location and geometry information. The texture refinement stage can adequately fuse such rich LF information to further improve LF reconstruction quality. Comprehensive experimental results validate the superiority of the proposed method. The rate-distortion performance and the perceptual quality of reconstructed views further demonstrate that the proposed method can save more bitrate while increasing LF reconstruction quality.

No abstract

No abstract

Light field (LF) imaging can obtain spatial and angular information of three-dimensional (3D) scene through a single shot, which enables a wide range of applications in the fields of 3D reconstruction, refocusing, virtual reality, etc. However, due to the inherent trade-off problem, the spatial resolution of acquired LF images is low, which hinders the widespread application of LF imaging technique. In order to relieve this issue, an end-to-end LF spatial super-resolution network is proposed by considering the multi-level perception and view reorganization. This method can fully explore the highly interwoven LF spatial and angular structure information. Specifically, a multi-feature fusion enhancement block is introduced that can fully perceive LF spatial, angular, and EPI information for LF spatial superresolution. Furthermore, the angular coherence between LF views is exploited by reorganizing the LF sub-aperture images and constructing a multi-angular stack structure. Compared with other state-of-the-art methods, the proposed method achieves superior performance in both visual and quantitative terms.