Zur Kurzanzeige

Semantic Self-adaptation: Enhancing Generalization with a Single Sample

dc.contributor.authorBahmani, Sherwin
dc.contributor.authorHahn, Oliver
dc.contributor.authorZamfir, Eduard
dc.contributor.authorAraslanov, Nikita
dc.contributor.authorCremers, Daniel
dc.contributor.authorRoth, Stefan
dc.date.accessioned2025-04-03T14:30:07Z
dc.date.available2025-04-03T14:30:07Z
dc.date.issued2023-07
dc.identifier.urihttps://tudatalib.ulb.tu-darmstadt.de/handle/tudatalib/4533
dc.descriptionThe 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.isoende_DE
dc.relationIsDescribedBy;arXiv;https://arxiv.org/abs/2208.05788
dc.relationIsDescribedBy;URL;https://openreview.net/forum?id=ILNqQhGbLx
dc.relationIsDescribedBy;URL;https://github.com/visinf/self-adaptive?tab=readme-ov-file
dc.rightsApache License 2.0
dc.rights.urihttps://www.apache.org/licenses/LICENSE-2.0
dc.subjectSemantic Segmentationde_DE
dc.subjectDomain Generalisationde_DE
dc.subjectDeep Learningde_DE
dc.subjectComputer Visionde_DE
dc.subject.classification4.43-04 Künstliche Intelligenz und Maschinelle Lernverfahrende_DE
dc.subject.classification4.43-05 Bild- und Sprachverarbeitung, Computergraphik und Visualisierung, Human Computer Interaction, Ubiquitous und Wearable Computingde_DE
dc.subject.ddc004
dc.titleSemantic Self-adaptation: Enhancing Generalization with a Single Samplede_DE
dc.typeSoftwarede_DE
tud.projectEC/H2020 | 866008 | REDde_DE
tud.projectHMWK | 500/10.001-(00012) | TAM - TP Rothde_DE
tud.unitTUDa


Dateien zu dieser Ressource

No Thumbnail [100%x60]
Name:
semantic_selfadapt_checkpoints.zip
Größe:
499.6MB
Format:
application/zip
Anzahl der Dateien
No Thumbnail [100%x60]
Name:
semantic_selfadapt_code.zip
Größe:
12.40MB
Format:
application/zip
Anzahl der Dateien
No Thumbnail [100%x60]
Name:
license_Apache-License.txt
Größe:
11.09KB
Format:
Text
Anzahl der Dateien

Der Datensatz erscheint in:

Zur Kurzanzeige

Apache License 2.0
Solange nicht anders angezeigt, wird die Lizenz wie folgt beschrieben: Apache License 2.0