Publications (10)
Color encoding in Latent Space of Stable Diffusion Models
Guillem Arias, Ariadna SolÃ, Martà Armengod +1
Recent advances in diffusion-based generative models have achieved remarkable visual fidelity, yet a detailed understanding of how specific perceptual attributes - such as color an…
Learning Relighting and Intrinsic Decomposition in Neural Radiance Fields
Yixiong Yang, Shilin Hu, Haoyu Wu +3
The task of extracting intrinsic components, such as reflectance and shading, from neural radiance fields is of growing interest. However, current methods largely focus on syntheti…
Deep intrinsic decomposition trained on surreal scenes yet with realistic light effects
Hassan Sial, Ramon Baldrich, Maria Vanrell
Estimation of intrinsic images still remains a challenging task due to weaknesses of ground-truth datasets, which either are too small or present non-realistic issues. On the other…
Relighting from a Single Image: Datasets and Deep Intrinsic-based Architecture
Yixiong Yang, Hassan Ahmed Sial, Ramon Baldrich +1
Single image scene relighting aims to generate a realistic new version of an input image so that it appears to be illuminated by a new target light condition. Although existing wor…
Light Direction and Color Estimation from Single Image with Deep Regression
Hassan A. Sial, Ramon Baldrich, Maria Vanrell +1
We present a method to estimate the direction and color of the scene light source from a single image. Our method is based on two main ideas: (a) we use a new synthetic dataset wit…
Understanding trained CNNs by indexing neuron selectivity
Ivet Rafegas, Maria Vanrell, Luis A. Alexandre +1
The impressive performance of Convolutional Neural Networks (CNNs) when solving different vision problems is shadowed by their black-box nature and our consequent lack of understan…
Intrinsic Decomposition of Document Images In-the-Wild
Sagnik Das, Hassan Ahmed Sial, Ke Ma +3
Automatic document content processing is affected by artifacts caused by the shape of the paper, non-uniform and diverse color of lighting conditions. Fully-supervised methods on r…
Color in Visual-Language Models: CLIP deficiencies
Guillem Arias, Ramon Baldrich, Maria Vanrell
This work explores how color is encoded in CLIP (Contrastive Language-Image Pre-training) which is currently the most influential VML (Visual Language model) in Artificial Intellig…
MLI-NeRF: Multi-Light Intrinsic-Aware Neural Radiance Fields
Yixiong Yang, Shilin Hu, Haoyu Wu +3
Current methods for extracting intrinsic image components, such as reflectance and shading, primarily rely on statistical priors. These methods focus mainly on simple synthetic sce…
Understanding learned CNN features through Filter Decoding with Substitution
Ivet Rafegas, Maria Vanrell
In parallel with the success of CNNs to solve vision problems, there is a growing interest in developing methodologies to understand and visualize the internal representations of t…