Haitao Zhou

Haitao Zhou (周海涛)

I am a Master's student from Beihang University(BUAA). And I am also an algorithm intern of AIsphere, where I work on video generation.

I got my Undergraduate's degree from the BUAA.

Recently, my research focuses on deep generative models(Video generation) and sketch&SVG generation.

Publications (* equal contribution, ^† corresponding author)

TrackGo: A Flexible and Efficient Method for Controllable Video Generation
Haitao Zhou*, Chuang Wang*, Rui Nie, Jinxiao Lin, Dnogdong Yu, Qian Yu^†, Changhu Wang^†
Association for the Advancement of Artificial Intelligence(AAAI), 2025

SVGDreamer++: Advancing Editability and Diversity in Text-Guided SVG Generation
Xingming Xing, Qian Yu^†, Chuang Wang, Haitao Zhou, Jing Zhang, Dong Xu
Transactions on Pattern Analysis and Machine Intelligence(T-PAMI), 2025

	SVGDreamer: Text Guided SVG Generation with Diffusion Model Xingming Xing, Haitao Zhou, Chuang Wang, Jing Zhang, Dong Xu, Qian Yu^† Computer Vision and Pattern Recognition(CVPR), 2024
	DiffSketcher: Text Guided Vector Sketch Synthesis through Latent Diffusion Models Xingming Xing, Chuang Wang, Haitao Zhou, Jing Zhang, Qian Yu^†, Dong Xu Neural Information Processing Systems (NeurIPS), 2023

Preprints (* equal contribution, ^† corresponding author)

	ViewCraft3D: High-Fidelity and View-Consistent 3D Vector Graphics Synthesis Chuang Wang, Haitao Zhou, Ling Luo^†, Qian Yu^† arXiv, 2025
	Inversion-by-Inversion: Exemplar-based Sketch-to-Photo Synthesis via Stochastic Differential Equations without Training Xingming Xing, Chuang Wang, Haitao Zhou, Zhihao Hu, Chongxuan Li, Dong Xu, Qian Yu^† arXiv, 2023

Projects

PyTorch-SVGRender

PyTorch-SVGRender is the go-to library for differentiable rendering methods in SVG generation. It supports a variety of vectorization methods, including: Image-to-SVG, Text-to-SVG, Text-to-Sketch.

Based on Pytorch, 2023