Multivariate Confidence Calibration for Object Detection
By Fabian Küppers, Jan Kronenberger, Amirhossein Shantia, Anselm Haselhoff. 2020
146 citations recorded by the source corpus.
Unbiased confidence estimates of neural networks are crucial especially for safety-critical applications. Many methods have been developed to calibrate biased confidence estimates. Though there is a variety of methods for classification, the field of object detection has not been addressed yet. Therefore, we present a novel framework to measure and calibrate biased (or miscalibrated) confidence estimates of object detection methods 1. The main difference to related work in the field of classifier calibration is that we also use additional information of the regression output of an object detector for calibration. Our approach allows, for the first time, to obtain calibrated confidence estimates with respect to image location and box scale. In addition, we propose a new measure to evaluate miscalibration of object detectors. Finally, we show that our developed methods outperform state-of-the-art calibration models for the task of object detection and provides reliable confidence estimates across different locations and scales.
Subfields: Artificial Intelligence
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
Citations in the corpus
- Task-Aware Information Routing from Common Representation Space in Lifelong Learning (2023, methodology): “Figure 5 shows the Expected Calibration Error (ECE) along with a reliability diagram on Seq-CIFAR100 using a calibration framework (Kuppers et al., 2020).”
- Task Agnostic Representation Consolidation: a Self-supervised based Continual Learning Approach (2022, methodology): “We report ECE along with a reliability diagram using a calibration framework (Küppers et al., 2020).”
- Consistency is the key to further mitigating catastrophic forgetting in continual learning (2022, methodology): “Figure 4 shows ECE along with a reliability diagram on S-CIFAR-10 using a calibration framework (Kuppers et al., 2020).”