EvCenterNet: Uncertainty Estimation for Object Detection using Evidential Learning
This addresses uncertainty estimation for object detection in safety-critical domains like automated driving, though it appears incremental as it builds on existing evidential learning and detection frameworks.
The authors tackled uncertainty estimation in 2D object detection for safety-critical applications like automated driving by proposing EvCenterNet, which uses evidential learning to estimate classification and regression uncertainties. Their approach improved precision and minimized execution time loss compared to the base model when evaluated on out-of-distribution datasets.
Uncertainty estimation is crucial in safety-critical settings such as automated driving as it provides valuable information for several downstream tasks including high-level decision making and path planning. In this work, we propose EvCenterNet, a novel uncertainty-aware 2D object detection framework using evidential learning to directly estimate both classification and regression uncertainties. To employ evidential learning for object detection, we devise a combination of evidential and focal loss functions for the sparse heatmap inputs. We introduce class-balanced weighting for regression and heatmap prediction to tackle the class imbalance encountered by evidential learning. Moreover, we propose a learning scheme to actively utilize the predicted heatmap uncertainties to improve the detection performance by focusing on the most uncertain points. We train our model on the KITTI dataset and evaluate it on challenging out-of-distribution datasets including BDD100K and nuImages. Our experiments demonstrate that our approach improves the precision and minimizes the execution time loss in relation to the base model.