New benchmark IPV-Bench evaluates image protection against AI video generation

Researchers have introduced IPV-Bench, a new benchmark designed to systematically evaluate image protection methods against image-to-video (I2V) generation. The benchmark includes a unified protocol for scoring protection effectiveness, visual fidelity, and robustness to preprocessing attacks, alongside IPV-500, a dataset covering five domains relevant to misuse scenarios. Experiments using IPV-Bench revealed that most protection methods offer limited disruption, protection rarely transfers between different I2V models, and simple preprocessing can break effective protection. AI

IMPACT Provides a standardized method for assessing the effectiveness of image protection techniques against AI-driven video generation.

RANK_REASON The cluster contains a research paper introducing a new benchmark for evaluating AI methods. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CV →

• Image-to-Video (I2V)
• IPV-500
• IPV-Bench
• Xiaofeng Li

• paper
• safety

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →