WE5.L4: Information Forensics and Security 2
Wed, 17 Sep, 16:30 - 18:00 Anchorage Time (UTC -8)
Location: Room 2
Session Type: Lecture
Session Chair: William Puech, Univ Montpellier
Track: [IF] Information Forensics and Security
Click the to view the manuscript on IEEE Xplore Open Preview
Wed, 17 Sep, 16:30 - 16:45 Anchorage Time (UTC -8)

WE5.L4.1: DIFFDEMORPH: EXTENDING REFERENCE-FREE DEMORPHING TO UNSEEN FACES
Nitish Shukla, Arun Ross, Michigan State University, United States
Wed, 17 Sep, 16:45 - 17:00 Anchorage Time (UTC -8)

WE5.L4.2: FACIAL IDENTITY EDITING: TOWARDS EFFECTIVE DE-IDENTIFICATION
Jinhyeong Park, Seangmin Lee, Muhammad Shaheryar, Soon Ki Jung, Kyungpook National University, Korea (South)
Wed, 17 Sep, 17:00 - 17:15 Anchorage Time (UTC -8)

WE5.L4.3: FACELIVT: FACE RECOGNITION USING LINEAR VISION TRANSFORMER WITH STRUCTURAL REPARAMETERIZATION FOR MOBILE DEVICE
Novendra Setyawan, National Formosa University, Taiwan; Chi-Chia Sun, National Taipei University, Taiwan; Mao-Hsiu Hsu, Wen-Kai Kuo, National Formosa University, Taiwan; Jun-Wei Hsieh, National Yang Ming Chiao Tung University, Taiwan
Wed, 17 Sep, 17:15 - 17:30 Anchorage Time (UTC -8)

WE5.L4.4: WaViT-CDC : Wavelet Vision Transformer with Central Difference Convolutions for Spatial-Frequency Deepfake Detection
Nour Eldin Alaa Badr, Jean-Christophe Nebel, Darrel Greenhill, Xing Liang, Kingston University School of Computer Science and Mathematics, United Kingdom of Great Britain and Northern Ireland
Wed, 17 Sep, 17:30 - 17:45 Anchorage Time (UTC -8)

WE5.L4.5: ENHANCING ADVERSARIAL ROBUSTNESS OF FOUNDATION MODELS WITHOUT DATA CENTRALIZATION
Koshiro Toishi, Keisuke Maeda, Ren Togo, Takahiro Ogawa, Miki Haseyama, Hokkaido University, Japan
Wed, 17 Sep, 17:45 - 18:00 Anchorage Time (UTC -8)

WE5.L4.6: FEW-SHOT CLASS-INCREMENTAL LEARNING FOR EFFICIENT SAR AUTOMATIC TARGET RECOGNITION
George Karantaidis, Athanasios Pantsios, Ioannis Kompatsiaris, Symeon Papadopoulos, Information Technologies Institute, Centre for Research and Technology Hellas, Greece