Rizhao Cai

Trustworthy AI Researcher

Huawei Singapore Research Center

I am a Trustworthy AI researcher at Huawei Singapore Research Center (since January 2025), working on EKYC security. I received my Ph.D. in 2024 from Nanyang Technological University (NTU), Singapore, supervised by Prof. Alex Kot at the ROSE Lab. Previously I served as a Research Project Manager in the ROSE Lab, leading industrial projects on neural network compression, object detection, image processing, and image generation.

Email Google Scholar GitHub LinkedIn ResearchGate

Research Interests

EKYC security: face anti-spoofing, deepfake detection, document recapture detection
Deep learning: OOD / domain generalization, continual learning

Selected Publications

View all

Towards Data-Centric Face Anti-Spoofing: Improving Cross-domain Generalization via Physics-based Data Synthesis

Rizhao Cai, C. Soh, Z. Yu, H. Li, W. Yang, A. Kot

International Journal of Computer Vision (IJCV), 2024

Code

BenchLMM: Benchmarking Cross-style Visual Capability of Large Multimodal Models

Rizhao Cai, Z. Song, D. Guan, Z. Chen, X. Luo, C. Yi, A. Kot

ECCV 2024, 2024 Oral

Project

S-Adapter: Generalizing Vision Transformer for Face Anti-Spoofing with Statistical Tokens

Rizhao Cai, Z. Yu, C. Kong, H. Li, C. Chen, Y. Hu, A. Kot

IEEE Transactions on Information Forensics and Security (T-IFS), 2024

Rehearsal-Free Domain Continual Face Anti-Spoofing: Generalize More and Forget Less

Rizhao Cai, Y. Cui, Z. Li, Z. Yu, H. Li, Y. Hu, A. Kot

ICCV 2023, 2023 Oral

Featured Projects

View all

FAS-Aug

Physics-based data synthesis for data-centric face anti-spoofing, improving cross-domain generalization.

Face Anti-SpoofingData Synthesis

BenchLMM

A benchmark for the cross-style visual capability of large multimodal models (ECCV 2024 Oral).

LMMBenchmark

MetaPattern_FAS

Learning meta patterns for domain-generalized face anti-spoofing (IEEE T-IFS 2022).

Face Anti-SpoofingDomain Generalization