AI RESEARCH

SynthForensics: Benchmarking and Evaluating People-Centric Synthetic Video Deepfakes

arXiv CS.CV

ArXi:2602.04939v2 Announce Type: replace Modern T2V/I2V generators synthesize people increasingly hard to distinguish from authentic footage, while current evaluation suites lag: legacy benchmarks target manipulation-based forgeries, and recent synthetic-video benchmarks prioritize scale over realistic human depiction. We