Calibrating Uncertainties in Object Localization Task
This work addresses a specific issue in uncertainty calibration for object localization, which is incremental as it applies an existing method to a new context.
The paper tackles the problem of uncalibrated uncertainty estimates in bounding box predictions for object localization, particularly in safety-critical applications like autonomous driving, by adapting an existing calibration technique for regression models, resulting in more reliable uncertainty estimates as demonstrated experimentally.
In many safety-critical applications such as autonomous driving and surgical robots, it is desirable to obtain prediction uncertainties from object detection modules to help support safe decision-making. Specifically, such modules need to estimate the probability of each predicted object in a given region and the confidence interval for its bounding box. While recent Bayesian deep learning methods provide a principled way to estimate this uncertainty, the estimates for the bounding boxes obtained using these methods are uncalibrated. In this paper, we address this problem for the single-object localization task by adapting an existing technique for calibrating regression models. We show, experimentally, that the resulting calibrated model obtains more reliable uncertainty estimates.