https://scholars.lib.ntu.edu.tw/handle/123456789/559407
標題: | Single-Image HDR Reconstruction by Learning to Reverse the Camera Pipeline | 作者: | Liu, Y.-L. Lai, W.-S. Chen, Y.-S. Kao, Y.-L. Yang, M.-H. Huang, J.-B. YUNG-YU CHUANG |
公開日期: | 2020 | 起(迄)頁: | 1648-1657 | 來源出版物: | Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition | 摘要: | Recovering a high dynamic range (HDR) image from a single low dynamic range (LDR) input image is challenging due to missing details in under-/over-exposed regions caused by quantization and saturation of camera sensors. In contrast to existing learning-based methods, our core idea is to incorporate the domain knowledge of the LDR image formation pipeline into our model. We model the HDR-to-LDR image formation pipeline as the (1) dynamic range clipping, (2) non-linear mapping from a camera response function, and (3) quantization. We then propose to learn three specialized CNNs to reverse these steps. By decomposing the problem into specific sub-tasks, we impose effective physical constraints to facilitate the training of individual sub-networks. Finally, we jointly fine-tune the entire model end-to-end to reduce error accumulation. With extensive quantitative and qualitative experiments on diverse image datasets, we demonstrate that the proposed method performs favorably against state-of-the-art single-image HDR reconstruction algorithms. © 2020 IEEE. |
URI: | https://www.scopus.com/inward/record.url?eid=2-s2.0-85093944674&partnerID=40&md5=6fc5e4a2b241120636909cf87e645972 https://scholars.lib.ntu.edu.tw/handle/123456789/559407 |
ISSN: | 10636919 | DOI: | 10.1109/CVPR42600.2020.00172 | SDG/關鍵字: | Cameras; Pattern recognition; Pipelines; Camera response functions; Error accumulation; High dynamic range images; Learning-based methods; Nonlinear mappings; Physical constraints; Qualitative experiments; Reconstruction algorithms; Image reconstruction |
顯示於: | 資訊工程學系 |
在 IR 系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。