CameraVDP: Perceptual Display Assessment with Uncertainty Estimation via Camera and Visual Difference Prediction

2025-09-10

AI Summary

Conventional display metrology struggles to accurately capture high-frequency or pixel-level spatial artifacts; while cameras offer high spatial resolution, they introduce optical, sampling, and photometric distortions, and physical measurements require integration with visual models to assess perceptual visibility of distortions. Method: We propose a perception-aware assessment framework that jointly models camera imaging and visual difference prediction. Our pipeline reconstructs fidelity-preserving images via HDR stacking, MTF deconvolution, vignetting and geometric correction, homography-based registration, and color calibration, then embeds a Visual Difference Predictor (VDP) to quantify human-perceivable distortions. Contribution/Results: We are the first to jointly model camera measurement uncertainty and VDP, deriving a theoretical upper bound on defect detection performance and outputting quality scores with confidence intervals. Validated on dead pixels, chromatic fringing, and luminance nonuniformity, our method achieves consistency between high-resolution measurement and human visual perception.

Technology Category

• Computer Vision: Adversarial Attacks & Robustness
• Machine Learning: Calibration & Uncertainty Quantification
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Security and Privacy: Large-scale security measurements
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Systems and Infrastructure for Web, Mobile and WoT: Web performance, measurement, and characterization

Abstract

Accurate measurement of images produced by electronic displays is critical for the evaluation of both traditional and computational displays. Traditional display measurement methods based on sparse radiometric sampling and fitting a model are inadequate for capturing spatially varying display artifacts, as they fail to capture high-frequency and pixel-level distortions. While cameras offer sufficient spatial resolution, they introduce optical, sampling, and photometric distortions. Furthermore, the physical measurement must be combined with a model of a visual system to assess whether the distortions are going to be visible. To enable perceptual assessment of displays, we propose a combination of a camera-based reconstruction pipeline with a visual difference predictor, which account for both the inaccuracy of camera measurements and visual difference prediction. The reconstruction pipeline combines HDR image stacking, MTF inversion, vignetting correction, geometric undistortion, homography transformation, and color correction, enabling cameras to function as precise display measurement instruments. By incorporating a Visual Difference Predictor (VDP), our system models the visibility of various stimuli under different viewing conditions for the human visual system. We validate the proposed CameraVDP framework through three applications: defective pixel detection, color fringing awareness, and display non-uniformity evaluation. Our uncertainty analysis framework enables the estimation of the theoretical upper bound for defect pixel detection performance and provides confidence intervals for VDP quality scores.

Problem

• Assessing perceptual display quality with camera-based measurements
• Addressing camera distortions for accurate display artifact detection
• Predicting visibility of display distortions under varying viewing conditions

Innovation

• Camera-based reconstruction pipeline with HDR stacking
• Visual Difference Predictor modeling human vision
• Uncertainty estimation framework with confidence intervals