Revisiting generalizability in deepfake detection: Improving metrics and stabilizing transfer

S Kamat, S Agarwal, T Darrell… - Proceedings of the …, 2023 - openaccess.thecvf.com
Abstract" Generalizability" is seen as the hallmark quality of a good deepfake detection
model. However, standard out-of-domain evaluation datasets are very similar in form to the …

Dagan++: Depth-aware generative adversarial network for talking head video generation

FT Hong, L Shen, D Xu - IEEE Transactions on Pattern Analysis …, 2023 - ieeexplore.ieee.org
Predominant techniques on talking head generation largely depend on 2D information,
including facial appearances and motions from input face images. Nevertheless, dense 3D …

Multimodal-driven talking face generation via a unified diffusion-based generator

C Xu, S Zhu, J Zhu, T Huang, J Zhang, Y Tai… - arxiv preprint arxiv …, 2023 - arxiv.org
Multimodal-driven talking face generation refers to animating a portrait with the given pose,
expression, and gaze transferred from the driving image and video, or estimated from the …

Make your actor talk: Generalizable and high-fidelity lip sync with motion and appearance disentanglement

R Yu, T He, A Zhang, Y Wang, J Guo, X Tan… - arxiv preprint arxiv …, 2024 - arxiv.org
We aim to edit the lip movements in talking video according to the given speech while
preserving the personal identity and visual details. The task can be decomposed into two …

Dreamhead: Learning spatial-temporal correspondence via hierarchical diffusion for audio-driven talking head synthesis

FT Hong, Y Liu, Y Li, C Zhou, F Yu, D Xu - arxiv preprint arxiv:2409.10281, 2024 - arxiv.org
Audio-driven talking head synthesis strives to generate lifelike video portraits from provided
audio. The diffusion model, recognized for its superior quality and robust generalization, has …

EAT-Face: Emotion-Controllable Audio-Driven Talking Face Generation via Diffusion Model

H Wang, X Jia, X Cao - 2024 IEEE 18th International …, 2024 - ieeexplore.ieee.org
Audio-driven talking face generation is a promising task with a lot of attention. Despite
abundant efforts are devoted to video quality and lip synchronization, most existing works do …