Publications

Multi-Shot Video Narrative.

Published in arXiv, 2026

This is a holistic framework for multi-shot video generation.

Download here

ViStoryBench: Comprehensive Benchmark Suite for Story Visualization

Published in CVPR 2026, 2026

This paper introduces ViStoryBench, a comprehensive benchmark designed to evaluate story visualization models across varied narrative structures, visual styles, and character settings.

Download here

StyleMe3D: Stylization with Disentangled Priors by Multiple Encoders on 3D Gaussians.

Published in arXiv, 2025

StyleMe3D is a holistic framework for 3D Gaussian Splatting style transfer that achieves scalable, semantically coherent, and perceptually enhanced stylization by disentangling multi-level semantics.