Improved image display by identifying the RGB family color space

2024-12-27

AI Summary

Color distortion in image display arises from unknown RGB color spaces. To address this, we propose an automatic color space identification method that jointly leverages pixel embedding and Gaussian process classification. Our approach is the first to integrate low-dimensional pixel embedding features with Gaussian processes for end-to-end discrimination among five major RGB gamuts—sRGB, Adobe RGB, Apple RGB, ColorMatch RGB, and ProPhoto RGB—without relying on image metadata or hand-crafted heuristics. Experiments on a diverse multi-source image dataset achieve 98.2% classification accuracy, substantially outperforming conventional rule-based methods. Moreover, we empirically demonstrate that accurate color space identification significantly improves display fidelity, validating its critical role in color management. The proposed framework provides a robust, deployable, and fully automated solution for practical color-space-aware imaging systems.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning

Application Category

• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Security and Privacy: Data transparency and provenance

Abstract

To display an image, the color space in which the image is encoded is assumed to be known. Unfortunately, this assumption is rarely realistic. In this paper, we propose to identify the color space of a given color image using pixel embedding and the Gaussian process. Five color spaces are supported, namely Adobe RGB, Apple RGB, ColorMatch RGB, ProPhoto RGB and sRGB. The results obtained show that this problem deserves more efforts.

Problem

• Color Optimization
• Image Display
• Color Setting

Innovation

• Color Profiling
• Pixel Information
• Mathematical Processes