3 papers
cs.CV2026
FreeSpec: Training-Free Long Video Generation via Singular-Spectrum Reconstruction
Fangda Chen, Shanshan Zhao, Longrong Yang +3
Video diffusion models perform well in short-video synthesis, but their training-free extension to long videos often suffers from content drift, temporal inconsistency, and over-sm…
cs.CV2025
JointTuner: Appearance-Motion Adaptive Joint Training for Customized Video Generation
Fangda Chen, Shanshan Zhao, Chuanfu Xu +1
Recent advancements in customized video generation have led to significant improvements in the simultaneous adaptation of appearance and motion. Typically, decoupling the appearanc…
cs.CV2025
"Stones from Other Hills can Polish Jade": Zero-shot Anomaly Image Synthesis via Cross-domain Anomaly Injection
Siqi Wang, Yuanze Hu, Xinwang Liu +5
Industrial image anomaly detection (IAD) is a pivotal topic with huge value. Due to anomaly's nature, real anomalies in a specific modern industrial domain (i.e. domain-specific an…