These tools and metrics are designed to help AI actors develop and use trustworthy AI systems and applications that respect human rights and are fair, transparent, explainable, robust, secure and safe.
nuScenes Detection Score (NDS)
The nuScenes Detection Score is a performance metric used to evaluate object detection algorithms in autonomous driving scenarios. It is used to evaluate the quality of object detection algorithms in the nuScenes dataset, which is a large-scale autonomous driving dataset.
NDS directly supports 'Safety' by quantifying how well an AI system detects objects, which is critical for preventing harm in autonomous driving scenarios. It also supports 'Robustness' by measuring detection performance across a variety of conditions and object types, indicating the system's reliability and resilience. The metric does not inherently address other Trustworthy AI objectives such as transparency, fairness, or privacy.
Trustworthy AI Relevance
This metric addresses Robustness by quantifying relevant system properties. NDS assesses detection accuracy and multiple error modes (position, orientation, velocity, etc.), which directly reflect an AI perception system's consistency and resilience to real-world variations (e.g., different object sizes, occlusion, motion). These properties map to the Trustworthy AI objective 'Robustness' because robustness concerns maintaining performance under adverse conditions and distribution shifts.
About the metric
You can click on the links to see the associated metrics
Objective(s):
Purpose(s):
Lifecycle stage(s):
Target users:
Partnership on AI
