Fraunhofer-Gesellschaft

Publica

Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Light-Field View Synthesis Using a Convolutional Block Attention Module

 
: Gul, Muhammad Shahzeb Khan; Mukati, M. Umair; Baetz, Michel; Forchhammer, Soren; Keinert, Joachim

:
Preprint urn:nbn:de:0011-n-6401944 (2.5 MByte PDF)
MD5 Fingerprint: 34e96505997f0b6c6ecbefbd88e8069b
© IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.
Erstellt am: 7.9.2021


Institute of Electrical and Electronics Engineers -IEEE-; IEEE Signal Processing Society:
IEEE International Conference on Image Processing, ICIP 2021 : 19-22 September 2021, Anchorage, Alaska, USA
Piscataway, NJ: IEEE, 2021
ISBN: 978-1-6654-4115-5
ISBN: 978-1-6654-3102-6
S.3398-3402
International Conference on Image Processing (ICIP) <2021, Anchorage/Alas.>
European Commission EC
H2020; 765911; RealVision
Hyperrealistic Imaging Experience
Englisch
Konferenzbeitrag, Elektronische Publikation
Fraunhofer IIS ()

Abstract
Consumer light-field (LF) cameras suffer from a low or limited resolution because of the angular-spatial tradeoff. To alleviate this drawback, we propose a novel learning-based approach utilizing attention mechanism to synthesize novel views of a light-field image using a sparse set of input views (i.e., 4 corner views) from a camera array. In the proposed method, we divide the process into three stages, stereo-feature extraction, disparity estimation, and final image refinement. We use three sequential convolutional neural networks for each stage. A residual convolutional block attention module (CBAM) is employed for final adaptive image refinement. Attention modules are helpful in learning and focusing more on the important features of the image and are thus sequentially applied in the channel and spatial dimensions. Experimental results show the robustness of the proposed method. Our proposed network outperforms the state-of-the-art learning-based light-field view synthesis methods on two challenging real-world datasets by 0.5 dB on average. Furthermore, we provide an ablation study to substantiate our findings.

: http://publica.fraunhofer.de/dokumente/N-640194.html