5 papers
Avatar-Forever: Decoupled Parallel Training for High-Quality Real-Time Infinite Avatars
Ruibin Li, Tao Yang, Zhiyuan Ma +3
Existing streaming video systems often rely on sequential, distillation-centered training pipelines to enable few-step long-video generation. However, this paradigm suffers from tw…
Long-Horizon Streaming Video Generation via Hybrid Attention with Decoupled Distillation
Ruibin Li, Tao Yang, Fangzhou Ai +4
Streaming video generation (SVG) distills a pretrained bidirectional video diffusion model into an autoregressive model equipped with sliding window attention (SWA). However, SWA i…
Many-for-Many: Unify the Training of Multiple Video and Image Generation and Manipulation Tasks
Ruibin Li, Tao Yang, Yangming Shi +4
Diffusion models have shown impressive performance in many visual generation and manipulation tasks. Many existing methods focus on training a model for a specific task, especially…
RORem: Training a Robust Object Remover with Human-in-the-Loop
Ruibin Li, Tao Yang, Song Guo +1
Despite the significant advancements, existing object removal methods struggle with incomplete removal, incorrect content synthesis and blurry synthesized regions, resulting in low…
Factorized-Dreamer: Training A High-Quality Video Generator with Limited and Low-Quality Data
Tao Yang, Yangming Shi, Yunwen Huang +3
Text-to-video (T2V) generation has gained significant attention due to its wide applications to video generation, editing, enhancement and translation, \etc. However, high-quality…