Resumen/Abstract
A color-naming model is a mechanism to categorize colors into a fixed number of basic color terms. These models are defined to be perceptual and categorize any pixel of the image into 11 different colors: red, green, blue, orange, brown, yellow, pink, purple, white, grey, and black. On the other hand, many editing tools offer a feature that allows the user to manipulate a limited selection of familiar colors. Editing by color name allows easy adjustment of elements like the "blue" of the sky or the "green" of trees. In this talk, we are motivated to study how to leverage this concept of color-name edition for different image enhancement tasks such as palette-based color harmonization [1], multiple image consistency [2], photographers' style editing [3], and image generation [4]. [1] Palette-Based Color Harmonization via Color Naming, Danna Xue, Javier Vazquez-Corral, Luis Herranz, and Yanning Zhang, Michael S. Brown, IEEE Signal Processing Letters, 2024 [2] Integrating High-Level Features for Consistent Palette-based Multi-image Recoloring, D. Xue, J. Vazquez-Corral, L. Herranz, and Y. Zhang, M. S. Brown, Computer Graphics Forum, 2023 [3] NamedCurves: Learned Image Enhancement via Color Naming, David Serrano-Lozano, Luis Herranz, Michael S Brown, and Javier Vazquez-Corral, European Conference on Computer Vision, 2024 [4] ColorPeel: Color Prompt Learning with Diffusion Models via Color and Shape Disentanglement, Muhammad Atif Butt, Kai Wang, Javier Vazquez-Corral, Joost Van De Weijer, European Conference on Computer Vision, 2024
Curriculum ponente
Dr. Javier Vazquez-Corral is a researcher in computational color with extensive teaching experience in many aspects of mathematics and computer science. He has developed novel approaches to solve different problems ranging from color constancy to color stabilization, color characterization, color gamut mapping, high dynamic range imaging, image dehazing, image denoising, and vision color properties such as unique hue prediction and color naming. He has over 70 publications, including 31 articles in top-ranked journals (26 as either first or second author), including IEEE-TPAMI, IEEE-TIP, PR, and SIAM-SIIMS. He has a European Patent granted -and a second application in process-, and research experience in cutting-edge research labs in Spain, the UK, Canada, and Switzerland, working on high-profile research projects. He has extensive experience teaching undergraduate and master courses in Spain and the UK, with more than 1200 hours of teaching already performed. Profesor proponente de la EPS: Marcos Escudero
Escuela Politécnica Superior | Universidad Autónoma de Madrid | Francisco Tomás y Valiente, 11 | 28049 Madrid | Tel.: +34 91 497 2222 | e-mail: informacion.eps@uam.es