Time-Dynamic Estimates of the Reliability of Deep Semantic Segmentation Networks
This work addresses safety-critical applications like automated driving by improving uncertainty estimation in semantic segmentation, though it appears incremental as it builds on existing uncertainty assessment methods by incorporating temporal dynamics.
The paper tackles the problem of assessing prediction reliability in deep semantic segmentation networks for street scenes by introducing a time-dynamic approach that tracks segments over time to gather aggregated metrics, enabling classification or prediction of intersection over union (IoU) values, with analysis on model performance and time series length effects.
In the semantic segmentation of street scenes with neural networks, the reliability of predictions is of highest interest. The assessment of neural networks by means of uncertainties is a common ansatz to prevent safety issues. As in applications like automated driving, video streams of images are available, we present a time-dynamic approach to investigating uncertainties and assessing the prediction quality of neural networks. We track segments over time and gather aggregated metrics per segment, thus obtaining time series of metrics from which we assess prediction quality. This is done by either classifying between intersection over union equal to 0 and greater than 0 or predicting the intersection over union directly. We study different models for these two tasks and analyze the influence of the time series length on the predictive power of our metrics.