Seeing enough: non-reference perceptual resolution selection for power-efficient client-side rendering

TL;DR

This paper introduces a non-reference method to select the minimal perceptual resolution for client-side rendering, reducing power consumption while maintaining visual quality on constrained devices.

Contribution

It proposes a novel, codec-agnostic neural approach that predicts perceptually indistinguishable resolutions without full-reference metrics, enabling power-efficient rendering.

Findings

Prediction improves perceptual quality with lower resolution.

Method reduces computational costs significantly.

Approach is applicable to various codecs and content types.

Abstract

Many client-side applications, especially games, render video at high resolution and frame rate on power-constrained devices, even when users perceive little or no benefit from all those extra pixels. Existing perceptual video quality metrics can indicate when a lower resolution is "good enough", but they are full-reference and computationally expensive, making them impractical for real-world applications and deployment on-device. In this work, we leverage the spatio-temporal limits of the human visual system and propose a non-reference method that predicts, from the rendered video alone, the lowest resolution that remains perceptually indistinguishable from the best available option, enabling power-efficient client-side rendering. Our approach is codec-agnostic and requires only minimal modifications to existing infrastructure. The network is trained on a large dataset of rendered content labeled with a full-reference perceptual video quality metric. The prediction significantly enhances perceptual quality while substantially reducing computational costs, suggesting a practical path toward perception-guided, power-efficient client-side rendering.