A. Generative Visual AI
14:00–15:40 (100 min) Venues: IPS S101
14:00-14:30 (30 min) Invited talk
Prof. Lizhuang Ma, Shanghai Jiao Tong University, China
- https://dmcv.sjtu.edu.cn/
- "Towards Generalizable Face Anti-Spoofing and Anti-Deepfake with Privacy"
- The metaverse heavily depends on biometric data to deliver immersive experiences. This study focuses on enhancing the adaptability and generalizability of anti-spoofing and deepfake detection systems in the metaverse. Our research contributes three innovative approaches.
1. Restricted Access Transfer via Source Model Reversion.
2. Efficient Domain Generalization with Adaptive Feature Learning.
3. Robust Facial Perception via Open-world and Continual Learning.
14:30-15:40 (70 min) Contribution papers
- 14:30-14:47 A1:"A Preliminary Diagnostic System for Motor Tics in Tourette Syndrome"
Asako Watanabe and Takashi Morishita (Fukuoka University)
- 14:47-15:04 A2:"Land Cover Classification for Multi-Band SAR Using Complex-Valued Cross-Attention Mechanism"
Kazuki Shiraishi, Sei-Ichiro Kamata (Waseda University)
- 15:04-15:21 A3:"Walking Trajectory Detection of Multiple Honeybees Using the Machine Learning Tool SLEAP"
Yuhao Jiang, Asako Watanabe, Shinya Takahashi and Hiroyuki Ai (Fukuoka University)
- 15:21-15:38 A4:"ADHD Diagnosis Using Spatial-Temporal Fractal Features from Key RSNs"
Mengyunqiu Zhang and Sei-ichiro Kamata (Waseda University)
