Oneta uses one transformer-based network with switchable style tokens and a ten-dimensional eigentransformation basis to run six enhancement tasks across 30 datasets, though it falls far behind specialists on dehazing and extreme low-light.
Semantic White Balance: Semantic Color Constancy Using Convolutional Neural Network
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
abstract
The goal of computational color constancy is to preserve the perceptive colors of objects under different lighting conditions by removing the effect of color casts caused by the scene's illumination. With the rapid development of deep learning based techniques, significant progress has been made in image semantic segmentation. In this work, we exploit the semantic information together with the color and spatial information of the input image in order to remove color casts. We train a convolutional neural network (CNN) model that learns to estimate the illuminant color and gamma correction parameters based on the semantic information of the given image. Experimental results show that feeding the CNN with the semantic information leads to a significant improvement in the results by reducing the error by more than 40%.
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Oneta: Multi-Style Image Enhancement Using Eigentransformation Functions
Oneta uses one transformer-based network with switchable style tokens and a ten-dimensional eigentransformation basis to run six enhancement tasks across 30 datasets, though it falls far behind specialists on dehazing and extreme low-light.