English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

The generalization with respect to domain shifts, as they frequently appearin applications such as autonomous driving, is one of the remaining bigchallenges for deep learning models. Therefore, we propose an exemplar-basedstyle synthesis pipeline to improve domain generalization in semanticsegmentation. Our method is based on a novel masked noise encoder for StyleGAN2inversion. The model learns to faithfully reconstruct the image, preserving itssemantic layout through noise prediction. Using the proposed masked noiseencoder to randomize style and content combinations in the training set, i.e.,intra-source style augmentation (ISSA) effectively increases the diversity oftraining data and reduces spurious correlation. As a result, we achieve up to$12.4\%$ mIoU improvements on driving-scene semantic segmentation underdifferent types of data shifts, i.e., changing geographic locations, adverseweather conditions, and day to night. ISSA is model-agnostic andstraightforwardly applicable with CNNs and Transformers. It is alsocomplementary to other domain generalization techniques, e.g., it improves therecent state-of-the-art solution RobustNet by $3\%$ mIoU in Cityscapes to DarkZ\"urich. In addition, we demonstrate the strong plug-n-play ability of theproposed style synthesis pipeline, which is readily usable for extra-sourceexemplars e.g., web-crawled images, without any retraining or fine-tuning.Moreover, we study a new use case to indicate neural network's generalizationcapability by building a stylized proxy validation set. This application hassignificant practical sense for selecting models to be deployed in theopen-world environment. Our code is available at\url{https://github.com/boschresearch/ISSA}.

Intra- & Extra-Source Exemplar-Based Style Synthesis for Improved Domain Generalization