Members of the Tsukada Laboratory attended the International Conference on Computer Vision (ICCV) 2025, held in Honolulu, Hawai‘i, from October 19th to 23rd, 2025. ICCV is one of the leading international conferences in the field of computer vision and brings together researchers from around the world.
Ph.D. student Hao Si presented a poster titled “You Share Beliefs, I Adapt: Progressive Heterogeneous Collaborative Perception.” The study examines the challenge of heterogeneity in collaborative perception for autonomous driving and proposes an approach that enables a vehicle model to adapt its parameters during inference without prior joint training.
Project Researcher Yue Yin attended several workshops related to visual localization, including Multi-Modal Localization and Mapping (MuLMa), 3D-VAST: From Street to Space, and Large-Scale Cross-Device Localization. Topics discussed included event camera–based SLAM, integration of foundation models with LiDAR and radar, scalable map representations, localization in dynamic environments, and privacy-aware localization techniques.
Laboratory members also joined the Workshop on Foundation Models for V2X-Based Cooperative Autonomous Driving, where recent efforts on incorporating large-scale pre-trained models into autonomous driving pipelines were introduced.
These activities provided valuable insights for the laboratory’s ongoing research on cooperative autonomous driving and intelligent transportation systems.
Publication
