革新知能統合研究センター

コンピュータビジョン分野のトップカンファレンスであるInternational Conference on Computer Visio (ICCV) 2025において、AIPセンターから6本の論文が採択されました。

[Website] https://iccv.thecvf.com/

[Accepted Papers]

• Disrupting Model Merging: A Parameter-Level Defense Without Sacrificing Accuracy
  Junhao Wei* (RIKEN AIP;Institute of Science Tokyo)
  Yu Zhe* (RIKEN AIP)
  Jun Sakuma (RIKEN AIP / Institute of Science Tokyo)
  * Equal contribution
• Embodied Navigation with Auxiliary Task of Action Description Prediction
  Haru Kondoh (Institute of Science Tokyo)
  Asako Kanezaki (Institute of Science Tokyo / RIKEN AIP)
• Frequency-Dynamic Attention Modulation For Dense Prediction
  Linwei Chen (Beijing Institute of Technology
  Lin Gu (RIKEN AIP)
  Ying Fu (Beijing Institute of Technology)
• GaussianOcc: Fully Self-supervised and Efficient 3D Occupancy Estimation with Gaussian Splatting
  Wanshui Gan (The University of Tokyo / RIKEN AIP)
  Fang Liu (The University of Tokyo)
  Hongbin Xu (South China University of Technology)
  Ningkai Mo (Deeproute)
  Naoto Yokoya (The University of Tokyo / RIKEN AIP)
• MP-HSIR: A Multi-Prompt Framework for Universal Hyperspectral Image Restoration
  Zhehui Wu (Wuhan University)
  Yong Chen (Jiangxi Normal University)
  Naoto Yokoya (The University of Tokyo / RIKEN AIP)
  Wei He (Wuhan University / RIKEN AIP)
• Transformed Low-rank Adaptation via Tensor Decomposition and Its Applications to Text-to-image Models
  Zerui Tao (RIKEN AIP)
  Yuhta Takida (Sony AI)
  Naoki Murata (Sony AI)
  Qibin Zhao (RIKEN AIP)
  Yuki Mitsufuji (Sony AI / Sony Group Corporation)

関連研究室