Reproduce “FlyVQA: Seeing Video Frame Interpolation Quality through Fly Eyes” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

Baoliang Chen, Lingyu Zhu, Zhiwei Zhong, Shiqi Wang

FlyVQA evaluates frame-interpolation quality by comparing reference and interpolated videos in a frozen, motion-trained fly visual circuit, including activity, excitatory and inhibitory inputs, and temporal coordination. On content-held-out BVI-VFI tests, it achieves Spearman and Pearson correlations of 0.733 and 0.738 with human ratings, above a retinal recurrent network. This supports interpolation assessment, not general video quality or human–fly neural alignment.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.