Flash photography enhancement via intrinsic relighting

By Elmar Eisemann, Frédo Durand. ACM Transactions on Graphics, 2004

684 citations recorded by the source corpus.

We enhance photographs shot in dark environments by combining a picture taken with the available light and one taken with the flash. We preserve the ambiance of the original lighting and insert the sharpness from the flash image. We use the bilateral filter to decompose the images into detail and large scale. We reconstruct the image using the large scale of the available lighting and the detail of the flash. We detect and correct flash shadows. This combines the advantages of available illumination and flash photography.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/eee03c81f38281479220ce323ee08e95ef84e541

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

Citations in the corpus

Deep Joint Image Filtering (2016, background): “For a target image, the guidance image can either be the target image itself [10,6], highresolution RGB images [6,2,3], images from different sensing modalities [11,12,5], or filtering outputs from previous iterations [9].”
Image deblurring with blurred/noisy image pairs (2007, methodology): “Denoising can be performed by a joint/cross bilateral filter using flash/no-flash images [Petschnigg et al. 2004; Eisemann and Durand 2004], or by an adaptive spatio-temporal accumulation filter for video sequences [Bennett and McMillan 2005].”
Local color transfer via probabilistic segmentation by expectation-maximization (2005, methodology): “In comparison, [4] used a compositing approach to inMeaning Gi Gaussian distribution Gi(i;μi, σi) for region i μi Mean of Gi at iteration t σ i Standard deviation of Gi at iteration t iP t xy Probability of pixel I(x, y) in Gi at iteration t Pxy Probability distribution of pixel I(x, y) estimated at”
Data-driven hallucination of different times of day from a single outdoor photo (2013, methodology): “Figure 11: We compare out model to the Joint Bilateral Filter [Eisemann and Durand 2004; Petschnigg et al. 2004] and the Guided Filter [He et al.”
Joint bilateral upsampling (2007, background): “…we have seen the introduction of joint (or cross ) bilateral ﬁlters in which the range ﬁlter is applied to a second guidance image , ˜ I , for example, when trying to combine the high frequencies from one image and the low frequencies from another [Petschnigg et al. 2004; Eisemann and Durand 2004].”
Two-scale tone management for photographic look (2006, methodology): “Strong edges are hard to characterize in |H|, which is why we define textureness using a crossbilateral filter [Eisemann and Durand 2004; Petschnigg et al. 2004] where the intensity image defines the edge-preserving term to filter |H|.”
LLNet: A deep autoencoder approach to natural low-light image enhancement (2015, methodology): “We approach the problem of low-light image enhancement from a representation learning perspective us-ing deep autoencoders (we refer to as Low-light Net, LL-Net) trained to learn underlying signal features in low-light images and adaptively brighten and denoise.”
A fusion-based enhancing method for weakly illuminated images (2016, background): “Other approaches include [24], where an automatic exposure correction is performed on image regions, and [25], where illumination from a naturally-lit image and details from a flash image are combined for image enhancement.”
Model Adaptation with Synthetic and Real Data for Semantic Dense Foggy Scene Understanding (2018, background): “However, as previous works on image ﬁltering [50] have shown, guided ﬁltering and similar joint ﬁltering methods such as cross-bilateral ﬁltering [14,44] transfer the structure that is present in the guidance/reference image to the output target image.”
Multi-Layer Background Subtraction Based on Color and Texture (2007, methodology): “To filter out noise, we propose to smooth the distance map using the cross bilateral filter introduced in [1].”
Weighted Guided Image Filtering (2016, background): “The BF is widely used due to its simplicity [9], [19], [20].”
A Fast Approximation of the Bilateral Filter Using a Signal Processing Approach (2006, methodology): “Finally, photography applications set the kernel spatial radius as a portion of the image size (Durand and Dorsey 2002; Eisemann and Durand 2004) and the range parameter can also be proportional to the intensity range (Petschnigg et al.”