N-dimensional probability density function transfer and its application to color transfer

By François Pitié, Anil C. Kokaram, Rozenn Dahyot. Tenth IEEE International Conference on Computer Vision (ICCV'05) Volume 1, 2005

400 citations recorded by the source corpus.

This article proposes an original method to estimate a continuous transformation that maps one N-dimensional distribution to another. The method is iterative, non-linear, and is shown to converge. Only 1D marginal distribution is used in the estimation process, hence involving low computation costs. As an illustration this mapping is applied to color transfer between two images of different contents. The paper also serves as a central focal point for collecting together the research activity in this area and relating it to the important problem of automated color grading

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/55f8eeae9077de2ef24c721b9dd8774cd8a2d151

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

Visual attribute transfer through deep image analogy (2017, background): “Early color transfer techniques tend to be global, i.e., a global transformation is applied to a source image to match the color statistics of a reference image [Reinhard et al. 2001; Pitie et al. 2005].”
Automated colour grading using colour distribution transfer (2007, background): “encompasses an entire range of possibilities from the simple match of the mean, variances [1], covariance [2,6] to the exact transfer of the whole probability density function (pdf) of the samples [7,3,5].”
Photorealistic Style Transfer via Wavelet Transforms (2019, methodology): “ photorealistic style transfer scenario as they are. Compared to the artistic style transfer, photorealistic transfer has been overlooked. The classical methods are mostly matching the color and tone [1,24,25] of the images, which are restricted to speciﬁc usage. Luan et al. [22] proposed deep photo”
Sliced and Radon Wasserstein Barycenters of Measures (2014, neutral): “D histogram matching used in [25] to perform color palette modiﬁcation.”
DoveNet: Deep Image Harmonization via Domain Verification (2019, methodology): “Image Harmonization: Traditional image harmonization methods concentrated on better matching low-level appearance statistics, such as matching global and local color distributions [35, 37], mapping to predefined perceptually harmonious color templates [4], applying gradient-domain compositing [34, 1”
Data-driven hallucination of different times of day from a single outdoor photo (2013, methodology): “Approaches for color transfer such as [Reinhard et al. 2001; Pouli and Reinhard 2011; Pitie et al. 2005] apply a global color mapping to match color statistics between images.”
Manipulating Attributes of Natural Scenes via Hallucination (2018, methodology): “In the past, statistical color transfer methods [Pitie et al. 2005; Reinhard et al. 2001] showed that the success of the style transfer methods highly depend on the semantic similarity of the source and target images.”
Deep Photo Style Transfer (2017, methodology): “Pitié’s histogram transfer [11], and Photoshop Match Color.”
Multivariate quantile mapping bias correction: an N-dimensional probability density function transform for climate model simulations of multiple variables (2017, methodology): “As noted above, a proof of convergence is provided by Pitié et al. (2007). Convergence to the target multivariate distribution is measured in the remainder of this paper using the energy distance (Székely and Rizzo 2004, 2013).”
A Closed-form Solution to Photorealistic Image Stylization (2018, methodology): “, color transfer [25, 24, 31] or tone transfer [1]) or are limited to specific scenarios (e.”
CG2Real: Improving the Realism of Computer Generated Images Using a Large Collection of Photographs (2011, methodology): “On the right are results from two color transfer algorithms: a global algorithm based on N-dimensional histogram matching [11], and our local color transfer algorithm.”
Enhancing Photorealism Enhancement (2021, methodology): “Subsequent work introduced increasingly sophisticated techniques for photographic style transfer between images [18], [48], [49], [50], [51], [52], [53], [54].”