Research
I publish as Chengze Li. I am an assistant professor at the School of Computing and Information Sciences, Saint Francis University. My research interests include Computational Comics and Manga, Computational Photography, and other fun stuff at the intersection of graphics and AI.
Featured Projects
See-through: Single-image Layer Decomposition for Anime Characters
ACM SIGGRAPH 2026 Conference Papers
Automatically decomposes a single anime illustration into fully inpainted, semantically distinct layers with inferred drawing order, turning static art into manipulatable 2.5D models.
ColorDiffuser: Video Colorization with Pretrained Text-to-Image Diffusion Models
ACM International Conference on Multimedia
Adapts a pretrained text-to-image latent diffusion model to video colorization, using Short- and Long-distance Cross-Frame Attention (SL-CFA) with amortized sampling for long-range temporal consistency, and a colorization VAE that curbs color bleeding with semantic boundaries and grayscale input.
Instance-guided Anime Editing with a Curated Large-scale Dataset
The Visual Computer
Introduces a 100k-image cartoon instance segmentation dataset and model, powering character-level editing applications like 3D Ken Burns parallax, style editing, and puppet animation.
A Contrastive Unified Encoding Framework for Sticker Style Editing
Pacific Conference on Computer Graphics and Applications
A contrastive learning framework that encodes each sticker pack's visual style into a unified latent space, enabling seamless style transfer, interpolation, and mixing across arbitrary sticker sets.
End-to-End Line Drawing Vectorization
AAAI Conference on Artificial Intelligence
Proposes an end-to-end Transformer-based method that directly converts raster line drawings to vector strokes in a single step, eliminating error-prone intermediate segmentation via automatic stroke-by-stroke tracing.
Seamless Manga Inpainting with Semantics Awareness
ACM Transactions on Graphics
Addresses manga inpainting after balloon/text removal by separating the task into semantic inpainting and appearance synthesis, with disentangled structural line and screentone features for better content understanding.
User-Guided Line Art Flat Filling With Split Filling Mechanism
IEEE/CVF Conference on Computer Vision and Pattern Recognition
A Split Filling Mechanism for user-guided line art flat filling that explicitly computes color scribble influence areas to prevent leakage between regions, while leveraging data-driven color generation learned from over a million illustrations.
Erasing Appearance Preservation in Optimization-Based Smoothing
European Conference on Computer Vision
Shows that partially erasing the appearance preservation term in optimization-based image smoothing facilitates better results, formulating automatic erasing as a 0-1 knapsack problem applicable to texture removal, intrinsic decomposition, and material editing.
Manga Filling Style Conversion with Screentone Variational Autoencoder
ACM Transactions on Graphics
Proposes ScreenVAE to map screened manga to an interpolative intermediate domain, unifying the properties of screentone and color filling for bidirectional translation between Western color comics and Japanese manga styles.
Two-stage Sketch Colorization
ACM Transactions on Graphics
A two-stage framework for sketch colorization: the drafting stage splashes varied colors, then the refinement stage detects and fixes artifacts like color distortion and blurring. Powers the popular style2paints tool.
Deep Extraction of Manga Structural Lines
ACM Transactions on Graphics
A CNN-based approach to extract structural lines from pattern-rich manga, suppressing arbitrary screentone patterns regardless of regularity, scale, or complexity, with no assumptions on pattern types.
All Publications
2026
Hanyuan Liu, Jinbo Xing, Minshan Xie, Chengze Li, Tien-Tsin Wong.
"Improved Diffusion-based Image Colorization via Piggybacked Models,"
IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPR Workshops), 2026.
Project
PDF (thecvf.com)
arXiv
Hao Xia, Xueting Liu, Chengze Li, Huisi Wu, Zhenkun Wen, Tong-Yee Lee.
"Integrating Visual and Contextual Cues for Ancient Chinese Rubbing Image Restoration,"
ACM Transactions on Multimedia Computing, Communications, and Applications (TOMM), 2026.
PDF (acm.org)
Jian Lin, Chengze Li, Haoyun Qin, Kwun Wang Chan, Yanghua Jin, Hanyuan Liu, Chun Wang Choy, Xueting Liu.
"See-through: Single-image Layer Decomposition for Anime Characters,"
ACM SIGGRAPH 2026 Conference Papers (SIGGRAPH), 2026.
Project
PDF (acm.org)
arXiv
Code
Xueting Liu, Shilong Deng, Chengze Li, Huisi Wu.
"Semantic-aware Multi-level Line Drawing Abstraction based on Line Importance Evaluation,"
International Conference on Image, Vision and Computing (ICIVC), 2026.
PDF (ieee.org)
Haoyun Qin, Chengze Li.
"Train Slit Photography: A Practical Pipeline,"
ACM SIGGRAPH Asia 2026 Posters (SIGGRAPH Asia Posters), 2026 (to appear).
2025
Minshan Xie, Jian Lin, Hanyuan Liu, Chengze Li, Tien-Tsin Wong.
"Advancing Manga Analysis: Comprehensive Segmentation Annotations for the Manga109 Dataset,"
IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2025.
PDF (cvf.com)
Dataset
Hanyuan Liu, Chengze Li, Minshan Xie, Zhenni Wang, Jiawen Liang, Chi-Sing Leung, Tien-Tsin Wong.
"BlueNeg: A 35mm Negative Film Dataset for Restoring Channel-Heterogeneous Deterioration,"
IEEE/CVF International Conference on Computer Vision (ICCV), 2025.
PDF (thecvf.com)
Dataset
Huisi Wu, Hao Meng, Chengze Li, Xueting Liu, Zhenkun Wen, Tong-Yee Lee.
"Cartoon Animation Outpainting With Region-Guided Motion Inference,"
IEEE Transactions on Visualization and Computer Graphics (TVCG), 2025.
PDF (ieee.org)
Zhenhua Ou, Chengze Li, Xueting Liu, Zhenkun Wen, Huisi Wu.
"Cartoon Animation Shading Removal,"
Computer Graphics International (CGI), 2025.
PDF (springer.com)
Hanyuan Liu, Chengze Li, Tien-Tsin Wong, Xueting Liu.
"ChromaFlow: End-to-End Flow Matching for Efficient Video Colorization,"
SIGGRAPH Asia Technical Communications (SA TC), 2025.
PDF (acm.org)
Hanyuan Liu, Minshan Xie, Jinbo Xing, Chengze Li, Chi-Sing Leung, Tien-Tsin Wong.
"ColorDiffuser: Video Colorization with Pretrained Text-to-Image Diffusion Models,"
ACM International Conference on Multimedia (ACM MM), 2025.
Project
PDF (acm.org)
arXiv
Jian Lin, Chengze Li, Xueting Liu, Zhongping Ge.
"Instance-guided Anime Editing with a Curated Large-scale Dataset,"
The Visual Computer (CGI), 2025.
Project
PDF (springer.com)
arXiv
Code
Dataset
Shilong Deng, Xueting Liu, Chengze Li, Ping Li, Zhenkun Wen, Huisi Wu.
"Line Drawing Abstraction Based on Line Importance Evaluation,"
Computer Graphics International (CGI), 2025.
PDF (springer.com)
Hao Xia, Xueting Liu, Chengze Li, Zhenkun Wen, Huisi Wu.
"Robust Character Stroke Segmentation for Diverse Fonts via Contour Matching and Chain Propagation,"
IEEE International Conference on Image Processing (ICIP), 2025.
PDF (ieee.org)
Minshan Xie, Menghan Xia, Chengze Li, Xueting Liu, Tien-Tsin Wong.
"Screentone-Preserved Manga Retargeting,"
Computer Graphics Forum (Eurographics), 2025.
PDF (wiley.com)
arXiv
Minshan Xie, Hanyuan Liu, Chengze Li, Tien-Tsin Wong.
"Synchronized Multi-Frame Diffusion for Temporally Consistent Video Stylization,"
Computer Graphics Forum (Eurographics), 2025.
PDF (wiley.com)
arXiv
Jian Lin, Chengze Li, Haoyun Qin, Hanyuan Liu, Xueting Liu, Xin Ma, Cunjian Chen, Tien-Tsin Wong.
"Trajectory-guided Anime Video Synthesis via Effective Motion Learning,"
Pacific Conference on Computer Graphics and Applications (Pacific Graphics), 2025.
Project
PDF (eg.org)
2024
Zhihong Ni, Chengze Li, Hanyuan Liu, Xueting Liu, Tien-Tsin Wong, Zhenkun Wen, Huisi Wu.
"A Contrastive Unified Encoding Framework for Sticker Style Editing,"
Pacific Conference on Computer Graphics and Applications (Pacific Graphics), 2024.
PDF (eg.org)
Wenpeng Xiao, Cheng Xu, Jiajie Mai, Xuemiao Xu, Yue Li, Chengze Li, Xueting Liu, Shengfeng He.
"Appearance-Preserved Portrait-to-Anime Translation via Proxy-Guided Domain Adaptation,"
IEEE Transactions on Visualization and Computer Graphics (TVCG), 2024.
PDF (ieee.org)
Zhenhua Ou, Xueting Liu, Chengze Li, Zhenkun Wen, Ping Li, Zhijian Gao, Huisi Wu.
"Body Part Segmentation of Anime Characters,"
Computer Animation and Virtual Worlds (CGI), 2024.
PDF (wiley.com)
Xueting Liu, Zibo Zhang, Hao Xia, Chengze Li, Huisi Wu.
"Calligraphic Guideline Generation via Contour-guided Brush Fitting,"
International Conference on Graphics and Image Processing (ICGIP), 2024.
PDF (spiedigitallibrary.org)
Haoyun Qin, Jian Lin, Hanyuan Liu, Xueting Liu, Chengze Li.
"Hyperstroke: A Novel High-quality Stroke Representation for Assistive Artistic Drawing,"
SIGGRAPH Asia Technical Communications (SA TC), 2024.
PDF (acm.org)
arXiv
Ziheng Ma, Chengze Li, Xueting Liu, Huisi Wu, Zhenkun Wen.
"Separating Shading and Reflectance From Cartoon Illustrations,"
IEEE Transactions on Visualization and Computer Graphics (TVCG), 2024.
PDF (ieee.org)
Huisi Wu, Ziheng Ma, Wenliang Wu, Xueting Liu, Chengze Li, Zhenkun Wen.
"Shading-Guided Manga Screening From Reference,"
IEEE Transactions on Visualization and Computer Graphics (TVCG), 2024.
PDF (ieee.org)
Jian Lin, Xueting Liu, Chengze Li, Minshan Xie, Tien-Tsin Wong.
"SKETCH2MANGA: Shaded Manga Screening from Sketch with Diffusion Models,"
IEEE International Conference on Image Processing (ICIP), 2024.
PDF (ieee.org)
arXiv
Code
2023
Yinghua Liu, Chengze Li, Xueting Liu, Huisi Wu, Zhenkun Wen.
"AddCR: a Data-driven Cartoon Remastering,"
The Visual Computer (CGI), 2023.
PDF (springer.com)
Xueting Liu, Chengze Li.
"Artificial Intelligence and Translation,"
Routledge Encyclopedia of Translation Technology, 2nd Edition (Book Chapter), 2023.
PDF (taylorfrancis.com)
Minshan Xie, Chengze Li, Tien-Tsin Wong.
"Manga Rescreening with Interpretable Screentone Representation,"
arXiv Preprint (arXiv), 2023.
arXiv
Chenshu Xu, Xuemiao Xu, Nanxuan Zhao, Weiwei Cai, Huaidong Zhang, Chengze Li, Xueting Liu.
"Panel-Page-Aware Comic Genre Understanding,"
IEEE Transactions on Image Processing (TIP), 2023.
PDF (ieee.org)
2022
Hanyuan Liu, Chengze Li, Xueting Liu, Tien-Tsin Wong.
"End-to-End Line Drawing Vectorization,"
AAAI Conference on Artificial Intelligence (AAAI), 2022.
Project
PDF (aaai.org)
Hanyuan Liu, Chengze Li, Xueting Liu, Tien-Tsin Wong.
"Neural Recognition of Dashed Curves with Gestalt Law of Continuity,"
IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2022.
Project
PDF (ieee.org)
Xueting Liu, Wenliang Wu, Chengze Li, Yifan Li, Huisi Wu.
"Reference-guided Structure-aware Deep Sketch Colorization for Cartoons,"
Computational Visual Media (CVM), 2022.
PDF (springer.com)
Zibo Zhang, Xueting Liu, Chengze Li, Huisi Wu, Zhenkun Wen.
"Shape-aware Stroke Segmentation for Calligraphic Characters,"
Computational Visual Media Conference (CVM), 2022.
Zibo Zhang, Xueting Liu, Chengze Li, Huisi Wu, Zhenkun Wen.
"Vectorizing Line Drawings of Arbitrary Thickness via Boundary-based Topology Reconstruction,"
Computer Graphics Forum (Eurographics), 2022.
PDF (wiley.com)
2021
Huisi Wu, Yifan Li, Xueting Liu, Chengze Li, Wenliang Wu.
"Deep Texture Cartoonization via Unsupervised Appearance Regularization,"
Computers & Graphics (CG), 2021.
PDF (elsevier.com)
Zhuming Zhang, Menghan Xia, Xueting Liu, Chengze Li, Tien-Tsin Wong.
"Invertible Tone Mapping with Selectable Styles,"
arXiv Preprint (arXiv), 2021.
arXiv
Lvmin Zhang, Chengze Li.
"Screenshots from Screen Photography,"
ACM SIGGRAPH Posters (SIGGRAPH Posters), 2021.
PDF (acm.org)
Minshan Xie, Menghan Xia, Xueting Liu, Chengze Li, Tien-Tsin Wong.
"Seamless Manga Inpainting with Semantics Awareness,"
ACM Transactions on Graphics (SIGGRAPH), 2021.
PDF (acm.org)
Code
Lvmin Zhang, Chengze Li, Edgar Simo-Serra, Yi Ji, Tien-Tsin Wong, Chunping Liu.
"User-Guided Line Art Flat Filling With Split Filling Mechanism,"
IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2021.
Project
PDF (cvf.com)
2020
Lvmin Zhang, Chengze Li, Yi Ji, Chunping Liu, Tien-Tsin Wong.
"Erasing Appearance Preservation in Optimization-Based Smoothing,"
European Conference on Computer Vision (ECCV), 2020.
Project
PDF (springer.com)
Code
Minshan Xie, Chengze Li, Xueting Liu, Tien-Tsin Wong.
"Manga Filling Style Conversion with Screentone Variational Autoencoder,"
ACM Transactions on Graphics (SIGGRAPH Asia), 2020.
PDF (acm.org)
Code
2019
Xinghong Hu, Xueting Liu, Zhuming Zhang, Menghan Xia, Chengze Li, Tien-Tsin Wong.
"Colorblind-shareable Videos by Synthesizing Temporal-coherent Polynomial Coefficients,"
ACM Transactions on Graphics (SIGGRAPH Asia), 2019.
Project
PDF (acm.org)
Yi Guo, Zhuming Zhang, Chu Han, Wenbo Hu, Chengze Li, Tien-Tsin Wong.
"Deep Line Drawing Vectorization via Line Subdivision and Topology Reconstruction,"
Computer Graphics Forum (Pacific Graphics), 2019.
PDF (wiley.com)
2018
Lvmin Zhang, Chengze Li, Tien-Tsin Wong, Yi Ji, Chunping Liu.
"Two-stage Sketch Colorization,"
ACM Transactions on Graphics (SIGGRAPH Asia), 2018.
Project
PDF (acm.org)
Code
2017
Xueting Liu, Chengze Li, Tien-Tsin Wong.
"Boundary-aware Texture Region Segmentation from Manga,"
Computational Visual Media (CVM), 2017.
PDF (springer.com)
Chengze Li, Xueting Liu, Tien-Tsin Wong.
"Deep Extraction of Manga Structural Lines,"
ACM Transactions on Graphics (SIGGRAPH), 2017.
PDF (acm.org)
Code
2015
Xueting Liu, Chengze Li, Haichao Zhu, Tien-Tsin Wong, Xuemiao Xu.
"Text-aware Balloon Extraction from Manga,"
The Visual Computer (CVM), 2015.
PDF (springer.com)
2014
Chengze Li, Xiaoyun Yuan, Juyong Zhang, Xiaoxin Lv, Lu Fang, Oscar C. Au.
"[Demo paper] Learning to Beautify Facial Image,"
IEEE International Conference on Multimedia and Expo Workshops (ICME Workshops), 2014.
PDF (ieee.org)