alt text  

Dr. Tianxiang Ma

Senior Algorithm Engineer, ByteDance
New Laboratory of Pattern Recognition
Institute of Automation, Chinese Academy of Sciences

Google Scholar  |  GitHub  |  Zhihu
Email: tianxiang.ma at cripac dot ia dot ac dot cn

Biography

Tianxiang Ma received the PhD degree from University of Chinese Academy of Sciences (UCAS) in 2024, under the supervision of Prof. Tieniu Tan and Prof. Jing Dong, and the B.Eng. degree from Zhengzhou University (ZZU) in 2019. Since July 2024, he has joined the Intelligent Creation Team, ByteDance Inc as a senior algorithm engineer. He interned at ByteDance Inc. and Alibaba DAMO Academy. His research focuses on computer vision and deep learning, particularly large-scale generative models for content creation, including Image, Video, and 3D Generation.

Personal Experience

alt text  

ByteDance Inc.       Jun. 2024 - Now
Senior Algorithm Engineer, Intelligent Creation Team


alt text  

University of Chinese Academy of Sciences (UCAS)       Sep. 2019 - Jun. 2024
PhD in Computer Applied Technology
Advisor: Prof. Tieniu Tan and Prof. Jing Dong


alt text  

Zhengzhou University (ZZU)       Sep. 2015 - Jun. 2019
B.Eng. in Electronic Science and Technology, Ranking: 1/93, GPA: 3.7/4.0
Advisors: Prof. Ronghui Luo and Dr. Yongzhi Tian

Selected Publications

alt text  

Phantom: Subject-Consistent Video Generation via Cross-Modal Alignment
Lijie Liu*, Tiangxiang Ma*, Bingchuan Li*, Zhuowei Chen*, Jiawei Liu, Qian He, Xinglong Wu
Arxiv, 2025
Paper | Project

alt text  

Freestyle 3D-Aware Portrait Synthesis Based on Compositional Generative Priors
Tiangxiang Ma, Kang Zhao, Jianxin Sun, Jing Dong, and Tieniu Tan
IEEE Conference on International Conference on Pattern Recognition (ICPR), 2025
Paper | Project

alt text  

GaFET: Learning Geometry-aware Facial Expression Translation from In-The-Wild Images
Tiangxiang Ma, Bingchuan Li, Qian He, Jing Dong, and Tieniu Tan
International Conference on Computer Vision (ICCV), 2023
Paper

alt text  

Semantic 3D-aware Portrait Synthesis and Manipulation Based on Compositional Neural Radiance Field
Tiangxiang Ma, Bingchuan Li, Qian He, Jing Dong, and Tieniu Tan
Proceedings of the AAAI Conference on Artificial Intelligence (AAAI), 2023 Oral
Paper | Project

alt text  

CFFT-GAN: Cross-domain Feature Fusion Transformer for Exemplar-based Image Translation
Tiangxiang Ma, Bingchuan Li, Wei Liu, Miao Hua, Jing Dong, and Tieniu Tan
Proceedings of the AAAI Conference on Artificial Intelligence (AAAI), 2023 Oral
Paper

alt text  

ReGANIE: Rectifying GAN Inversion Errors for Accurate Real Image Editing
Bingchuan Li, Tiangxiang Ma, Peng Zhang, Miao Hua, Wei Liu, Qian He, and Zili Yi
Proceedings of the AAAI Conference on Artificial Intelligence (AAAI), 2023 Oral
Paper

alt text  

AdaDeId: Adjust Your Identity Attribute Freely
Tiangxiang Ma, Dongze Li, Wei Wang, and Jing Dong
IEEE Conference on International Conference on Pattern Recognition (ICPR), 2022
Paper

alt text  

CFA-Net: Controllable Face Anonymization Network with Identity Representation Manipulation
Tiangxiang Ma, Dongze Li, Wei Wang, and Jing Dong
arXiv
Paper

alt text  

MUST-GAN: Multi-level Statistics Transfer for Self-driven Person Image Generation
Tiangxiang Ma, Bo Peng, Wei Wang, and Jing Dong
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2021
Paper | Project

alt text  

Any-to-one Face Reenactment Based on Conditional Generative Adversarial Network
Tiangxiang Ma, Bo Peng, Wei Wang, and Jing Dong
Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC), 2019
Paper

Selected Honors and Awards

Academic Services

Journal and Conference Reviewer