Search Results for author:
Found 5 papers, 2 papers with code
Efficient Video Face Enhancement with Enhanced Spatial-Temporal Consistency
In Stage \Rmnum{1}, we learn the model with a regularizer mitigating the codebook collapse problem.
EchoMimicV2: Towards Striking, Simplified, and Semi-Body Human Animation
Recent work on human animation usually involves audio, pose, or movement maps conditions, thereby achieves vivid animation quality.
Token Caching for Diffusion Transformer Acceleration
Diffusion transformers have gained substantial interest in diffusion generative modeling due to their outstanding performance.
EchoMimic: Lifelike Audio-Driven Portrait Animations through Editable Landmark Conditions
Through the implementation of a novel training strategy, EchoMimic is capable of generating portrait videos not only by audios and facial landmarks individually, but also by a combination of both audios and selected facial landmarks.
A Novel DeBERTa-based Model for Financial Question Answering Task
As a rising star in the field of natural language processing, question answering systems (Q&A Systems) are widely used in all walks of life.
Cannot find the paper you are looking for? You can
Submit a new open access paper.
