3 papers
cs.CV2026
VideoMemory: Toward Consistent Video Generation via Memory Integration
Jinsong Zhou, Yihua Du, Xinli Xu +7
Maintaining consistent characters, props, and environments across multiple shots is a central challenge in narrative video generation. Existing models can produce high-quality shor…
cs.CV2025
Multi-level Dynamic Style Transfer for NeRFs
Zesheng Li, Shuaibo Li, Wei Ma +2
As the application of neural radiance fields (NeRFs) in various 3D vision tasks continues to expand, numerous NeRF-based style transfer techniques have been developed. However, exi…
cs.CV2025
TrueMoE: Dual-Routing Mixture of Discriminative Experts for Synthetic Image Detection
Laixin Zhang, Shuaibo Li, Wei Ma +1
The rapid progress of generative models has made synthetic image detection an increasingly critical task. Most existing approaches attempt to construct a single, universal discrimi…