2486–2490 (2018)Įl Helou, M., Zhou, R., Barthas, J., Süsstrunk, S.: VIDIT: virtual image dataset for illumination transfer. In: IEEE International Conference on Image Processing (ICIP), pp. Image and Video Processing arXiv:2004.08541 (2020)Įl Helou, M., Dümbgen, F., Süsstrunk, S.: AAM: an assessment metric of axial chromatic aberration. Sabarinathan, Beham, M., Roomi, S.: Moire image restoration using multi level hyper vision net. In: Proceedings of the European Conference on Computer Vision Workshops (ECCVW) (2020)ĭ. arXiv preprint arXiv:2006.02333 (2020)ĭong, L., Jiang, Z., Li, C.: An ensemble neural network for scene relighting with light classification. 248–255 (2009)ĭherse, A.P., Everaert, M.N., Gwizdała, J.J.: Scene relighting with illumination estimation in the latent space on an encoder-decoder scheme. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. (TOG) 35(6), 1–8 (2016)ĭeng, J., Dong, W., Socher, R., Li, L.J., Li, K., Fei-Fei, L.: ImageNet: a large-scale hierarchical image database.
arXiv preprint arXiv:2001.10773 (2020)Ĭhen, J., Adams, A., Wadhwa, N., Hasinoff, S.W.: Bilateral guided upsampling. 1–10 (2009)īurton, G.J., Moorhead, I.R.: Color and spatial structure in natural scenes. 57–70 (2012)īell, S., Bala, K., Snavely, N.: Intrinsic images in the wild. In: European Conference on Computer Vision (ECCV), pp. 379–387 (2015)īarron, J.T., Malik, J.: Color constancy, intrinsic images, and shape estimation. In: IEEE International Conference on Computer Vision (ICCV), pp.
243–252 (2019)Īfifi, M., Price, B., Cohen, S., Brown, M.S.: When color constancy goes wrong: correcting improperly white-balanced images. 11 (2019)Īfifi, M., Brown, M.S.: What else can fool deep learning? addressing color constancy errors on deep neural network performance. In: British Machine Vision Conference (BMVC), p. KeywordsĪfifi, M., Brown, M.S.: Sensor-independent illumination estimation for DNN models.
The tracks had 94, 52, and 56 registered participants, respectively, leading to 20 confirmed submissions in the final competition stage. Participants were allowed to make use of their track 1 and 2 solutions for track 3. The target color temperature and orientation, rather than being pre-determined, are instead given by a guide image. Lastly, the third track dealt with any-to-any relighting, thus a generalization of the first track. The goal of the second track was to estimate illumination settings, namely the color temperature and orientation, from a given image. The first track considered one-to-one relighting the objective was to relight an input photo of a scene with a different color temperature and illuminant orientation (i.e., light source position). This paper presents the novel VIDIT dataset used in the challenge and the different proposed solutions and final evaluation results over the 3 challenge tracks. We review the AIM 2020 challenge on virtual image relighting and illumination estimation.