Content-Insensitive Dynamic Lip Feature Extraction for Visual Speaker Authentication Against Deepfake Attacks.

Zihao Guo, Shilin Wang

Browse the full ICASSP paper archive.