Seeing Beyond 8bits: Subjective and Objective Quality Assessment of HDR-UGC Videos

TL;DR

This paper introduces a large-scale HDR-UGC video quality dataset and a novel multimodal language model, HDR-Q, that effectively assesses HDR video quality by leveraging HDR-specific features and reasoning, outperforming existing models.

Contribution

The work presents the first HDR-aware vision encoder and a specialized RL finetuning framework, HAPO, for HDR-UGC video quality assessment, advancing beyond SDR-based models.

Findings

HDR-Q achieves state-of-the-art performance on HDR-VQA benchmarks.

The Beyond8Bits dataset contains 44K videos with over 1.5 million crowd ratings.

HDR-Q effectively captures HDR-specific distortions and quality cues.

Abstract

High Dynamic Range (HDR) user-generated (UGC) videos are rapidly proliferating across social platforms, yet most perceptual video quality assessment (VQA) systems remain tailored to Standard Dynamic Range (SDR). HDR has a higher bit depth, wide color gamut, and elevated luminance range, exposing distortions such as near-black crushing, highlight clipping, banding, and exposure flicker that amplify UGC artifacts and challenge SDR models. To catalyze progress, we curate Beyond8Bits, a large-scale subjective dataset of 44K videos from 6.5K sources with over 1.5M crowd ratings, spanning diverse scenes, capture conditions, and compression settings. We further introduce HDR-Q, the first Multimodal Large Language Model (MLLM) for HDR-UGC VQA. We propose (i) a novel HDR-aware vision encoder to produce HDR-sensitive embeddings, and (ii) HDR-Aware Policy Optimization (HAPO), an RL finetuning framework that anchors reasoning to HDR cues. HAPO augments GRPO via an HDR-SDR contrastive KL that encourages token reliance on HDR inputs and a Gaussian weighted regression reward for fine-grained MOS calibration. Across Beyond8Bits and public HDR-VQA benchmarks, HDR-Q delivers state-of-the-art performance.