What Do Audio-Visual Synchronization Metrics Actually Measure?
Not on the map. Why?
Abstract (arXiv)
Automatic AV-sync metrics are widely used to rank and train audio-visual generators, but they are rarely audited as measurement instruments. We jointly audit AV-Align, ImageBind AV-relevance, JavisScore, and Synchformer/DeSync under a common reliability protocol: controlled-distortion monotonicity, preprocessing sensitivity, rank uncertainty, cross-metric agreement, PEAVS-proxy agreement, and learned fusion. The result is an axis split, not a single winner: Synchformer/DeSync is the strongest temporal-offset tracker (τ=0.84), ImageBind/JavisScore better match the PEAVS human-aligned proxy (τ=0.20) and content-disruption families, and AV-Align is the weakest standalone metric. The metrics mutually disagree (Krippendorff α=0.066), and neither linear nor simple k-NN fusion improves PEAVS agreement over the best individual metric. We recommend reporting AV-sync as a Reliability Card (metric-family breakdowns with confidence intervals) rather than a single bare synchronization score.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 10 of its 12 references that are in the map. First posted on arXiv 25 Aug 2026.
Builds on
10 of 12 references.
References · 10
- 2023ImageBind: One Embedding Space To Bind Them Allcited by 1,175
- 2024MMAudio: Taming Multimodal Joint Training for High-Quality Video-to-Audio Synthesiscited by 134
- 2024Synchformer: Efficient Synchronization from Sparse Cuescited by 97
- 2025JavisDiT: Joint Audio-Video Diffusion Transformer with Hierarchical Spatio-Temporal Prior Synchronizationcited by 88
- 2023Diverse and Aligned Audio-to-Video Generation via Text-to-Video Model Adaptationcited by 69
- 2025VABench: A Comprehensive Benchmark for Audio-Video Generationcited by 30
- 2024PEAVS: Perceptual Evaluation of Audio-Visual Synchrony Grounded in Viewers' Opinion Scorescited by 13
- 2024STREAM: Spatio-TempoRal Evaluation and Analysis Metric for Video Generative Modelscited by 6
- 2026AVBench: Human-Aligned and Automated Evaluation Benchmark for Audio-Video Generative Modelscited by 5
- 2026SLVMEval: Synthetic Meta Evaluation Benchmark for Text-to-Long Video Generationcited by 2
Built on by
Not cited yet.