Runtime Monitoring for Out-of-Distribution Detection in Object Detection Neural Networks
| Autoři | |
|---|---|
| Rok publikování | 2023 |
| Druh | Stať ve sborníku |
| Konference | FORMAL METHODS, FM 2023 |
| Fakulta / Pracoviště MU | |
| Citace | |
| Doi | https://doi.org/10.1007/978-3-031-27481-7_36 |
| Klíčová slova | Runtime monitoring; Neural networks; Out-of-distribution detection; Object detection |
| Popis | Runtime monitoring provides a more realistic and applicable alternative to verification in the setting of real neural networks used in industry. It is particularly useful for detecting out-of-distribution (OOD) inputs, for which the network was not trained and can yield erroneous results. We extend a runtime-monitoring approach previously proposed for classification networks to perception systems capable of identification and localization of multiple objects. Furthermore, we analyze its adequacy experimentally on different kinds of OOD settings, documenting the overall efficacy of our approach. |
| Související projekty: |