Objective, Absolute and Hue-aware Metrics for Intrinsic Image Decomposition on Real-World Scenes: A Proof of Concept

2025-05-26

AI Summary

Quantitative evaluation of intrinsic image decomposition (IID) in real-world scenarios remains challenging due to the absence of ground-truth albedo and shading. Existing annotation-based metrics suffer from subjectivity, relativity, and insensitivity to hue. To address these limitations, this work proposes the first objective, absolute, and hue-aware quantitative evaluation paradigm for IID. We introduce physically interpretable computational albedo—derived from hyperspectral imaging and LiDAR intensity modeling—as a reference standard. A spectral-similarity-driven albedo densification method enables absolute-scale quantification while preserving hue fidelity. Experimental validation in controlled laboratory settings confirms that the proposed metric achieves objectivity, absolute scale consistency, and explicit hue awareness. This framework establishes a reproducible and generalizable theoretical and technical foundation for evaluating IID performance in realistic conditions.

Technology Category

• Computer Vision: Interpretability, Explainability, and Transparency
• Machine Learning: Calibration & Uncertainty Quantification
• Humans and AI: Other Foundations of Human Computation & AI

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets

Abstract

Intrinsic image decomposition (IID) is the task of separating an image into albedo and shade. In real-world scenes, it is difficult to quantitatively assess IID quality due to the unavailability of ground truth. The existing method provides the relative reflection intensities based on human-judged annotations. However, these annotations have challenges in subjectivity, relative evaluation, and hue non-assessment. To address these, we propose a concept of quantitative evaluation with a calculated albedo from a hyperspectral imaging and light detection and ranging (LiDAR) intensity. Additionally, we introduce an optional albedo densification approach based on spectral similarity. This paper conducted a concept verification in a laboratory environment, and suggested the feasibility of an objective, absolute, and hue-aware assessment. (This paper is accepted by IEEE ICIP 2025. )

Problem

• Lack of ground truth for IID quality assessment
• Subjectivity in human-judged albedo annotations
• Need for hue-aware evaluation in real-world scenes

Innovation

• Uses hyperspectral imaging for albedo calculation
• Employs LiDAR intensity for quantitative evaluation
• Introduces spectral similarity-based albedo densification