2025
CVPR Highlight
Prototypical skeleton action recognition, Hongda Liu et al.

NLPR · MAIS · CASIA
The Smart Identity Recognition group studies human-centric biometrics: iris and ocular imaging, light-field cameras, privacy-preserving templates, and perception when a person will not stand still.
Led by Prof. Tieniu Tan and Prof. Zhenan Sun, with faculty Yunlong Wang and Kunbo Zhang.
The group
SIR is part of the New Laboratory of Pattern Recognition and the State Key Laboratory of Multimodal Artificial Intelligence Systems at the Institute of Automation, Chinese Academy of Sciences. MAIS was approved in 2022, when the pattern-recognition and complex-systems national key laboratories were reorganized into one of the first benchmark national key laboratories.
The work covers imaging hardware, acquisition, preprocessing, feature encoding, matching, and the privacy of the template that comes out. Iris, sclera, periocular, face, three-dimensional human perception, and millimeter-wave sensing are all in scope, with a standing focus on non-cooperative recognition at a distance.
Featured
Recognition
2025
Prototypical skeleton action recognition, Hongda Liu et al.
2025
Yunlong Wang, October 2025.
2024
Two federated-learning patents, 22 October 2024.
2023
Sclera-TransFuse, presented to Haiqing Li.
2022
Second prize, Technology Invention Award of the China Society of Image and Graphics.
2020
All-in-Focus Iris Camera with a Great Capture Volume.
Gazette
Research

Iris
The full iris pipeline, from segmentation and encoding to recognition when the eye is moving, distant, or only partly open.

Ocular
Sclera, periocular skin, and fused ocular regions for identity when a full iris texture is not available.

Light field
Microlens imaging for refocus, depth, liveness, and iris capture across a much larger volume.

Imaging
Optics, sensors, and capture software for iris on the move, including a multi-person system at 10 meters.

Trust
Privacy, federated learning, adversarial defense, and demographic coverage for biometric systems that have to be believed.

Perception
Bodies, hands, gait-scale motion, and millimeter-wave sensing when the face is not the only signal.
2026
Yanting Wu, Qi Li, Yunlong Wang, Fangling Jiang et al.
IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
2026
Haoyan Jiang, Siqi Guo, Yunlong Wang, Caiyong Wang
Electronics
2026
Hongda Liu, Yunfan Liu, Min Ren, Lin Su et al.
IEEE Transactions on Biometrics, Behavior, and Identity Science (TBIOM)
2026
Leyuan Wang, Liuyu Xiang, Yujie Wei, Yiwei Ru et al.
IEEE Transactions on Image Processing (TIP)