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

Using multiple spatial modalities has been proven helpful in improvingsemantic segmentation performance. However, there are several real-worldchallenges that have yet to be addressed: (a) improving label efficiency and(b) enhancing robustness in realistic scenarios where modalities are missing atthe test time. To address these challenges, we first propose a simple yetefficient multi-modal fusion mechanism Linear Fusion, that performs better thanthe state-of-the-art multi-modal models even with limited supervision. Second,we propose M3L: Multi-modal Teacher for Masked Modality Learning, asemi-supervised framework that not only improves the multi-modal performancebut also makes the model robust to the realistic missing modality scenariousing unlabeled data. We create the first benchmark for semi-supervisedmulti-modal semantic segmentation and also report the robustness to missingmodalities. Our proposal shows an absolute improvement of up to 10% on robustmIoU above the most competitive baselines. Our code is available athttps://github.com/harshm121/M3L

Missing Modality Robustness in Semi-Supervised Multi-Modal Semantic Segmentation