Safety Metric
Ken Yu's safety metric research for self-driving perception and risk-aware scene understanding.
Summary
This research project explores safety metrics for self-driving perception and scene understanding. Ken Yu studied how to quantify risk in traffic scenes so perception outputs can be evaluated not only for accuracy, but also for their impact on safety-critical decisions.
Focus areas: scene understanding, safety-aware evaluation, autonomous driving perception, metric design.
Motivation
Related Work
- Panoptic-DepthLab for richer scene understanding with depth.
- Perspective-aware Convolution for monocular 3D object detection.
- Publications for related academic output.
Introduction
Result
