Local Temperature Scaling for Probability Calibration

Abstract

For semantic segmentation, label probabilities are often uncalibrated as they are typically only the by-product of a segmentation task. Intersection over Union (IoU) and Dice score are often used as criteria for segmentation success, while metrics related to label probabilities are rarely explored. On the other hand, probability calibration approaches have been studied, which aim at matching probability outputs with experimentally observed errors, but they mainly focus on classification tasks, not on semantic segmentation. Thus, we propose a learning-based calibration method that focuses on multi-label semantic segmentation. Specifically, we adopt a tree-like convolution neural network to predict local temperature values for probability calibration. One advantage of our approach is that it does not change prediction accuracy, hence allowing for calibration as a post-processing step. Experiments on the COCO and LPBA40 datasets demonstrate improved calibration performance over different metrics. We also demonstrate the performance of our method for multi-atlas brain segmentation from magnetic resonance images.

Publication
2021 IEEE/CVF International Conference on Computer Vision, ICCV 2021, Montreal, QC, Canada, October 10-17, 2021
Zhipeng Ding
Zhipeng Ding
Ph.D. in Computer Science

My research interests include image registration and machine learning.

Xu Han
Xu Han
Ph.D. in Computer Science

My research interests include image registration for images with pathologies and machine learning.

Peirong Liu
Peirong Liu
Graduate Student in Computer Science

My research interests include machine learning, computer vision, partial differential equations, and perfusion imaging.

Marc Niethammer
Marc Niethammer
Professor of Computer Science

My research interests include image registration, image segmentation, shape analysis, machine learning, and biomedical applications.

Related