I am an AI Engineer working on multi-modal learning and digital human generation, with a focus on speech-driven facial generation, digital human synthesis, and AIGC generation & understanding. I have strong experience in Diffusion Models, Flow Matching , DiT, VQ-VAE/GAN, ControlNet, LoRA, IP-Adapter, Stable Diffusion, FLUX and so on.

Previously, I worked at Xiaobing, where I developed digital human and talking-face generation systems, also contributed to real-time avatar rendering.

I also worked as an Algorithm Engineer at Orbbec and Moqi, companies focusing on 3D sensing and biometrics.

I received my Master’s degree from the School of Computer Science, Northwestern Polytechnical University (NWPU), Xi’an, China, under the supervision of Prof. WeiWei and Prof. ZhangLei.

I received my Bachelor’s degree in the school of computer science, NorthWestern Polytechnical University(NWPU), Xi’an, China.

Publications

1. IConFace: Identity-Structure Asymmetric Conditioning for Unified Reference-Aware Face Restoration. Axi Niu, Jinyang Zhang, Senyan Qing. arXiv:2605.02814, 2026. Project / Code Coming Soon
2. The Second Challenge on Real-World Face Restoration at NTIRE 2026: Methods and Results. Jingkai Wang et al. NTIRE 2026 organizers and challenge participants, including Axi Niu, Jinyang Zhang, Zhenguo Wu, and Senyan Qing. arXiv:2604.10532v2 / CVPR 2026 Workshop, 2026.
3. Improving Hyperspectral Image Classification with Unsupervised Knowledge Learning. Jinyang Zhang, Wei Wei, Lei Zhang, et al. IEEE International Geoscience and Remote Sensing Symposium (IGARSS). IEEE, 2019: 2722-2725. Code
4. Learning Discriminative Compact Representation for Hyperspectral Imagery Classification. Lei Zhang, Jinyang Zhang, Wei Wei, et al. IEEE Transactions on Geoscience and Remote Sensing (TGRS), 2019, 57(10): 8276-8289. Code
5. Deep Cube-Pair Network for Hyperspectral Imagery Classification. Wei Wei, Jinyang Zhang, Lei Zhang, et al. Remote Sensing, 2018, 10(5): 783. Code
6. Data-specific Activation Function Learning for Hyperspectral Image Classification. Yu Li, Wei Wei, Jinyang Zhang, Lei Zhang, et al. International Conference on Orange Technology (ICOT), 2019.
7. Fast-convergent fully connected deep learning model using constrained nodes input. Chen Ding, Ying Li, Lei Zhang, Jinyang Zhang, Lu Yang, et al. Neural Processing Letters, 2019, 49(3): 995-1005.
8. Boosting Hyperspectral Image Classification With Unsupervised Feature Learning. Wei Wei, Songzheng Xu, Lei Zhang, Jinyang Zhang, Yanning Zhang. IEEE Transactions on Geoscience and Remote Sensing (TGRS), 2021, 0196-2892.

Honors and Awards

• Outstanding Graduate Student of 2020, School of Computer Science, Northwestern Polytechnical University

For more info

More information can be found in my CV.