The Box Size Confidence Bias Harms Your Object Detector

Johannes Gilg, Torben Teepe, Fabian Herzog, Gerhard Rigoll; Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), 2023, pp. 1471-1480

Abstract


Countless applications depend on accurate predictions with reliable confidence estimates from modern object detectors. However, it is well known that neural networks, including object detectors, produce miscalibrated confidence estimates. Recent work even suggests that detectors' confidence predictions are biased with respect to object size and position. In object detection, the issues of conditional biases, confidence calibration, and task performance are usually explored in isolation, but, as we aim to show, they are closely related. We formally prove that the conditional confidence bias harms the performance of object detectors and empirically validate these findings. Specifically, to quantify the performance impact of the confidence bias on object detectors, we modify the histogram binning calibration to avoid performance impairment and instead improve it through calibration conditioned on the bounding box size. We further find that the confidence bias is also present in detections generated on the training data of the detector, which can be leveraged to perform the de-biasing. Moreover, we show that Test Time Augmentation (TTA) confounds this bias, which results in even more significant performance impairments on the detectors. Finally, we use our proposed algorithm to analyze a diverse set of object detection architectures and show that the conditional confidence bias harms their performance by up to 0.6 mAP and 0.8 mAP50. Code available at https://github.com/Blueblue4/Object-Detection-Confidence-Bias.

Related Material


[pdf] [supp] [arXiv]
[bibtex]
@InProceedings{Gilg_2023_WACV, author = {Gilg, Johannes and Teepe, Torben and Herzog, Fabian and Rigoll, Gerhard}, title = {The Box Size Confidence Bias Harms Your Object Detector}, booktitle = {Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)}, month = {January}, year = {2023}, pages = {1471-1480} }