HubLens › Trending › bilibili/Index-anisora
bilibili

Index-anisora

AIMachine LearningGenerative AIVideo GenerationComputer VisionDeep Learning
View on GitHub
2,402
+340

// summary

Index-AniSora is a powerful open-source framework designed specifically for high-quality anime video generation. The system integrates a comprehensive data processing pipeline, a controllable generation model with spatiotemporal masking, and a specialized evaluation benchmark. It supports diverse creative applications including character 3D generation, video style transfer, and multimodal guidance.

// use cases

01
Character 3D video generation from front-facing illustrations
02
Video style transfer using line art and frame-based guidance
03
Precise motion control via multimodal guidance such as pose, depth, and audio