Perception that reads the world
Robust 2D, 3D, panoramic, and multimodal scene understanding for complex real-world environments.
Research architecture
InSAI studies intelligent systems that can sense, understand, navigate, and support people in the physical world. The work spans computer vision, robotics, transportation, multimodal learning, and accessible interaction.
Robust 2D, 3D, panoramic, and multimodal scene understanding for complex real-world environments.
Embodied intelligence, co-navigation, instruction following, and safe human-AI collaboration.
Sensor fusion, semantic mapping, and resilient decision-making for vehicles and robotic systems.
Assistive interfaces, wearable systems, document accessibility, and inclusive human-computer interaction.

The Assistive Agents for All workshop solicits research that explores AI agents supporting people in everyday life - particular, research that empowers users while preserving agency, safety, privacy, accessibility, fairness, and accountability. Please check our workshop webpage at: https://insailab.org/ubicomp2026-aaa-workshop/.

The 1st iCARE workshop @CoRL2025 will provide a half-day session (poster, oral, invited talks) to integrate insights from assistive technology and cobodied AI into the robotics and learning community. Please check our workshop webpage at: https://insailab.org/icare/.

The workshop "Beyond Supervised Learning: addressing data scarcity in Intelligent Transportation Systems" brings together autonomous driving and computer vision researchers and practitioners from industry and academia. Please check our workshop webpage at: https://sites.google.com/view/iv2022-bsl-workshop.