高動態範圍（ HDR）影像旨在捕獲現實場景的所有亮度範圍，但受限於拍攝之硬體設備與顯示器， HDR 影像時常無法良好 地 顯示於一般消費 級顯示器上， 假如將高動態範圍影像於顯示裝置上顯示，就必須做動態範圍的映射。 目前最常見的 HDR 影像重建多是使用多曝光進行曝光融合技術，但缺點在於若對齊並未處理好，則生成的照片可能會產生殘影或偽影的現象，故本篇論文著重使用單曝光重建 HDR 之技術。基於 U-net 結合區域權重來偵測飽和區域在對過曝區域進行更深入的處理。

High Dynamic Range (HDR) images are designed to capture the full range of brightness of the real scene. Limited by the hardware used for shooting and the display, HDR images often cannot be displayed well on general consumer monitors. Display on the device, it is necessary to do dynamic range mapping. At present, the most common HDR image reconstruction is to use multi-exposure for exposure fusion technology, but the disadvantage is that if the alignment is not processed well, the generated photos may have artifacts, so this paper focuses on using single exposure technique of reconstructing HDR. Detecting saturated regions based on U-net combined with local weighting module aims to perform more in-depth processing of over-exposure regions.

[1] G. Eilertsen, J. Kronander, G. Denes, R. K. Mantiuk, and J. Unger, “HDR image reconstruction from a single exposure using deep CNNS,” in ACM Trans. Graph., vol. 36, no. 6, p. 178, 2017.
[2] X. Chen, Y. Liu, Z. Zhang, Y. Qiao and C. Dong, “HDRUNet: Single Image HDR Reconstruction with Denoising and Dequantization,” in Proc. IEEE/CVF Conf. CVPR, p. 354 – 363, 2021.
[3] P. E. Debevec, J. Malik, “Recovering high dynamic range radiance maps from photographs,” in SIGGRAPH, p.369 – 378, Aug. 1997.
[4] S. Mann and R. W. Picard, “On being ‘undigital’ with digital cameras: Extending dynamic range by combining differently exposed pictures,” in Proceedings of IS&T, 1995.
[5] N. K. Kalantari and R. Ramamoorthi, “Deep High Dynamic Range Imaging of Dynamic Scenes,” in ACM Trans. Graph., vol. 36, no. 4, 2017.
[6] S. Wu, J. Xu, Y. W. Tai, and C. K. Tang, “Deep High Dynamic Range Imaging with Large Foreground Motions,” in CV – ECCV, p. 120 – 135, 2018
[7] F. Bouzaraa, I. Halfaoui and O. Urfalioglu, “Learnable Exposure Fusion for Dynamic Scenes,” in CVPR, Apr. 2018.
[8] F. Banterle, K. Debattista, A. Artusi, S. Pattanaik, K. Myszkowski, P. Ledda, and A. Chalmers, “High dynamic range imaging and low dynamic range expansion for generating HDR content,” in Computer Graphics Forum, 2009.
[9] F. Banterle, P. Ledda, K. Debattista, and A. Chalmers, “Inverse tone mapping. In International conference,” on Computer graphics and interactive techniques in Australasia and Southeast Asia, 2006.
[10] F. Banterle, P. Ledda, K. Debattista, and A. Chalmers, “Expanding low dynamic range videos for high dynamic range applications,” in Spring Conference on Computer Graphics, 2008.
[11] F. Banterle, P. Ledda, K. Debattista, and A. Chalmers, and M. Bloj, “A framework for inverse tone mapping,” in The Visual Computer, 2007.
[12] Y. Huo, F. Yang, L. Dong, and V. Brost, “Physiological inverse tone mapping based on retina response,” in The Visual Computer, 2014.
[13] R. P Kovaleski and M. M Oliveira, “High-quality reverse tone mapping for a wide range of exposures,” in 2014 27th SIBGRAPI Conference on Graphics, Patterns and Images, 2014.
[14] 17. G. Eilertsen, J. Kronander, G. Denes, R. K. Mantiuk, and J. Unger, “HDR image reconstruction from a single exposure using deep CNNS,” in ACM Trans. Graph., vol. 36, no. 6, p. 178, 2017.
[15] Y. L. Liu, W. S. Lai, Y. S. Chen, Y. L. Kao, M. H. Yang, Y. Y. Chuang, and J. B. Huang, “Single-image HDR reconstruction by learning to reverse the camera pipeline,'' in Proc. IEEE/CVF Conf. CVPR, p. 1651-1660, Jun. 2020.
[16] D. Marnerides, T. Bashford-Rogers, J. Hatchett and K. Debattista, “ExpandNet: A Deep Convolutional Neural Network for High Dynamic Range Expansion from Low Dynamic Range Content,” in CVPR, 2018.
[17] S. Lin, J. Gu, S. Yamazaki, and H. Y. Shum, “Radiometric calibration from a single image,” in CVPR, 2004.
[18] S. Lin and L. Zhang, “Determining the radiometric response function from a single grayscale image,” in CVPR, 2005.
[19] A. Odena, V. Dumoulin, C. Olah, “Deconvolution and checkerboard artifacts. Distill,”, Distill, 2016.
[20] S. Iizuka, E. Simo-Serra, H. Ishikawa, “Let there be Color!: Joint End-to-end Learning of Global and Local Image Priors for Automatic Image Colorization with Simultaneous Classification,” in ACM Trans. Graph., 2016.
[21] S. Ioffe, C. Szegedy, “Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift,” in ICML, 2015.
[22] A. Krizhevsky, I. Sutskever, G. E. Hinton, “Imagenet classification with deep convolutional neural networks,” in Commun. ACM 60, 6, p. 84–90, May, 2017.
[23] M.Mathieu, C. Couprie, Y. LeCun, “Deep multi-scale video prediction beyond mean square error,” abs/1511.05440, 2015.
[24] S. K. Nayar and T. Mitsunaga, “High dynamic range imaging: Spatially varying pixel exposures,” in Proc. IEEE Conf. CVPR, p. 472–479, Jun. 2000.
[25] J. Tumblin, A. Agrawal, and R. Raskar, “Why i want a gradient camera,” in Proc. IEEE Comput. Soc. Conf. CVPR, pp. 103–110, Jun. 2005.
[26] R. Timofte, V. D. Smet, and L. V. Gool, “A+: Adjusted anchored neighborhood regression for fast super-resolution,” in Asian Conference on Computer Vision, p. 111–126, 2014.
[27] S. Gu, N. Sang, and F. Ma, “Fast image super resolution via local regression,” in IEEE Conference on International Conference on Pattern Recognition, p. 3128–3131, 2012.
[28] P. Isola, J. Y. Zhu, T. Zhou, A. A. Efros, “Image-to-Image Translation with Conditional Adversarial Networks,” in IEEE CVPR, 2017.
[29] Z. Khan and M. Khanna and S. Raman, “FHDR: HDR Image Reconstruction from a Single LDR Image using Feedback Network,” in GlobalSIP, p.1-5, 2019.
[30] M. S. Santos, T. I. Ren, N. K. Kalantari, “Single Image HDR Reconstruction Using a CNN with Masked Features and Perceptual Loss, ” in ACM Trans. Graph., vol. 39, no. 4, Jul. 2020.
[31] H. Yu, W. Liu, C. Long, B. Dong, Q. Zou, C. Xiao, “Luminance Attentive Networks for HDR Image and Panorama Reconstruction,” in arXiv:2109.06688, 2021.