1. 연사: 김산민 (국민대학교/자동차IT융합학과)

2. 주제: 자율주행을 위한 이미지 기반 3차원 인지 기술(Vision-based 3D Perception for Autonomous Driving)

3. 일시: 5/28 (목요일), 오후 4시~5시 30분

4. 장소: L409

5. Abstract : Cameras offer rich semantic cues and cost-effective scalability, making vision-based 3D perception a central pillar of modern autonomous driving. This talk briefly reviews the evolution of image-based 3D object detection along spatial and temporal axes, together with training strategies that mitigate the inherent depth ambiguity of cameras, and 3D occupancy prediction as a complementary scene representation. Building on this foundation, the talk then argues that the shift toward end-to-end autonomous driving demands a fundamental rethinking of perception itself. Rather than an isolated module, perception should be planning-aware and reactive, integrating feedback from prediction and planning to dynamically adapt perceptual saliency to the driving context, and evaluated by planning-centric metrics that directly reflect downstream driving performance.