Human Action CLIPs: Detecting AI-generated Human Motion
This addresses the need for protection against AI-generated video misuse, though it appears incremental as it builds on existing detection methods with a new dataset.
The paper tackles the problem of distinguishing real from AI-generated human motion to protect against malicious uses, achieving effective detection using multi-modal semantic embeddings that are robust to resolution and compression attacks.
AI-generated video generation continues its journey through the uncanny valley to produce content that is increasingly perceptually indistinguishable from reality. To better protect individuals, organizations, and societies from its malicious applications, we describe an effective and robust technique for distinguishing real from AI-generated human motion using multi-modal semantic embeddings. Our method is robust to the types of laundering that typically confound more low- to mid-level approaches, including resolution and compression attacks. This method is evaluated against DeepAction, a custom-built, open-sourced dataset of video clips with human actions generated by seven text-to-video AI models and matching real footage. The dataset is available under an academic license at https://www.huggingface.co/datasets/faridlab/deepaction_v1.