EC-IoU: Orienting Safety for Object Detectors via Ego-Centric Intersection-over-Union
This addresses safety concerns in autonomous navigation by refining evaluation metrics, though it is incremental as it builds on existing IoU methods.
The paper tackles the limitation of standard IoU in evaluating safety for object detectors in navigation by proposing EC-IoU, a weighting mechanism that prioritizes closer points from the ego agent's perspective, and shows it improves mean Average Precision on the KITTI dataset.
This paper presents Ego-Centric Intersection-over-Union (EC-IoU), addressing the limitation of the standard IoU measure in characterizing safety-related performance for object detectors in navigating contexts. Concretely, we propose a weighting mechanism to refine IoU, allowing it to assign a higher score to a prediction that covers closer points of a ground-truth object from the ego agent's perspective. The proposed EC-IoU measure can be used in typical evaluation processes to select object detectors with better safety-related performance for downstream tasks. It can also be integrated into common loss functions for model fine-tuning. While geared towards safety, our experiment with the KITTI dataset demonstrates the performance of a model trained on EC-IoU can be better than that of a variant trained on IoU in terms of mean Average Precision as well.