Abstract | ||
---|---|---|
Employing representations generated by large-scale training in a transfer-learning setting achieves state-of-the-art anomaly segmentation results when applied to the visual inspection task. Current approaches, however, focus exclusively on features of pre-trained classification networks, which are known to posess lower spatial resolution than segmentation or object detection networks. In our work, we investigate whether features extracted from pre-trained segmentation networks can be used to further improve anomaly segmentation performance in the transfer-learning setting. To this end, we apply stateof-the-art transfer-learning methods to encoder-decoder based segmentation networks. Results show that the encoders of pre-trained segmentation networks yield improved anomaly segmentation performance compared to their pre-trained classification counterparts. However, no consistent improvements can be observed yet regarding the decoders of the pre-trained segmentation networks. Together, this demonstrates that pre-trained segmentation networks can be used to further improve transfer-learned anomaly segmentation performance and that additional research is required to fully unleash their potential. |
Year | DOI | Venue |
---|---|---|
2021 | 10.1109/ETFA45728.2021.9613387 | 2021 26TH IEEE INTERNATIONAL CONFERENCE ON EMERGING TECHNOLOGIES AND FACTORY AUTOMATION (ETFA) |
Keywords | DocType | ISSN |
Anomaly Segmentation, Transfer Learning, Probability Density Estimation, Gaussian Density Estimation | Conference | 1946-0740 |
Citations | PageRank | References |
0 | 0.34 | 0 |
Authors | ||
2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Oliver Rippel | 1 | 0 | 1.35 |
Dorit Merhof | 2 | 0 | 2.37 |