Exploring Opinion-unaware Video Quality Assessment with Semantic Affinity Criterion
This addresses the problem of expensive and biased video quality assessment for applications requiring scalable and reliable evaluation without human input, though it is incremental as it builds on existing low-level metrics.
The paper tackles the high cost and bias of human opinion-dependent video quality assessment (VQA) by proposing an opinion-unaware method that incorporates high-level semantics using CLIP-based semantic affinity, combined with low-level naturalness metrics. The result is a 20% improvement over existing opinion-unaware VQA methods and increased robustness compared to opinion-aware approaches.
Recent learning-based video quality assessment (VQA) algorithms are expensive to implement due to the cost of data collection of human quality opinions, and are less robust across various scenarios due to the biases of these opinions. This motivates our exploration on opinion-unaware (a.k.a zero-shot) VQA approaches. Existing approaches only considers low-level naturalness in spatial or temporal domain, without considering impacts from high-level semantics. In this work, we introduce an explicit semantic affinity index for opinion-unaware VQA using text-prompts in the contrastive language-image pre-training (CLIP) model. We also aggregate it with different traditional low-level naturalness indexes through gaussian normalization and sigmoid rescaling strategies. Composed of aggregated semantic and technical metrics, the proposed Blind Unified Opinion-Unaware Video Quality Index via Semantic and Technical Metric Aggregation (BUONA-VISTA) outperforms existing opinion-unaware VQA methods by at least 20% improvements, and is more robust than opinion-aware approaches.