Semantic Self-adaptation: Enhancing Generalization with a Single Sample
| dc.contributor.author | Bahmani, Sherwin | |
| dc.contributor.author | Hahn, Oliver | |
| dc.contributor.author | Zamfir, Eduard | |
| dc.contributor.author | Araslanov, Nikita | |
| dc.contributor.author | Cremers, Daniel | |
| dc.contributor.author | Roth, Stefan | |
| dc.date.accessioned | 2025-04-03T14:30:07Z | |
| dc.date.available | 2025-04-03T14:30:07Z | |
| dc.date.issued | 2023-07 | |
| dc.identifier.uri | https://tudatalib.ulb.tu-darmstadt.de/handle/tudatalib/4533 | |
| dc.description | The lack of out-of-domain generalization is a critical weakness of deep networks for semantic segmentation. Previous studies relied on the assumption of a static model, i. e., once the training process is complete, model parameters remain fixed at test time. In this work, we challenge this premise with a self-adaptive approach for semantic segmentation that adjusts the inference process to each input sample. Self-adaptation operates on two levels. First, it fine-tunes the parameters of convolutional layers to the input image using consistency regularization. Second, in Batch Normalization layers, self-adaptation interpolates between the training and the reference distribution derived from a single test sample. Despite both techniques being well known in the literature, their combination sets new state-of-the-art accuracy on synthetic-to-real generalization benchmarks. Our empirical study suggests that self-adaptation may complement the established practice of model regularization at training time for improving deep network generalization to out-of-domain data. | de_DE |
| dc.language.iso | en | de_DE |
| dc.relation | IsDescribedBy;arXiv;https://arxiv.org/abs/2208.05788 | |
| dc.relation | IsDescribedBy;URL;https://openreview.net/forum?id=ILNqQhGbLx | |
| dc.relation | IsDescribedBy;URL;https://github.com/visinf/self-adaptive?tab=readme-ov-file | |
| dc.rights | Apache License 2.0 | |
| dc.rights.uri | https://www.apache.org/licenses/LICENSE-2.0 | |
| dc.subject | Semantic Segmentation | de_DE |
| dc.subject | Domain Generalisation | de_DE |
| dc.subject | Deep Learning | de_DE |
| dc.subject | Computer Vision | de_DE |
| dc.subject.classification | 4.43-04 Künstliche Intelligenz und Maschinelle Lernverfahren | de_DE |
| dc.subject.classification | 4.43-05 Bild- und Sprachverarbeitung, Computergraphik und Visualisierung, Human Computer Interaction, Ubiquitous und Wearable Computing | de_DE |
| dc.subject.ddc | 004 | |
| dc.title | Semantic Self-adaptation: Enhancing Generalization with a Single Sample | de_DE |
| dc.type | Software | de_DE |
| tud.project | EC/H2020 | 866008 | RED | de_DE |
| tud.project | HMWK | 500/10.001-(00012) | TAM - TP Roth | de_DE |
| tud.unit | TUDa |
Files in this item
This item appears in the following Collection(s)
-
Segmentation [9]
Except where otherwise noted, this item's license is described as Apache License 2.0