To address the problems of difficult feature extraction, small differences between texture information and recognition accuracy to be improved in lip-print recognition tasks, a lip-print recognition algorithm based on grouped multi-scale feature fusion is proposed. The ablation experimental results show that the improved recognition model achieves 98.56% recognition accuracy on the test set, and the model has strong generalization ability and feature refinement expression ability, which can provide support for the application of lip-print recognition technology in the field of identity verification.