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/

Zendy Plus

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

Zendy Tools

Zendy Open

Multi-task dense prediction aims at handling multiple pixel-wise predictiontasks within a unified network simultaneously for visual scene understanding.However, cross-task feature interactions of current methods are still sufferingfrom incomplete levels of representations, less discriminative semantics infeature participants, and inefficient pair-wise task interaction processes. Totackle these under-explored issues, we propose a novel BridgeNet framework,which extracts comprehensive and discriminative intermediate Bridge Features,and conducts interactions based on them. Specifically, a Task PatternPropagation (TPP) module is firstly applied to ensure highly semantictask-specific feature participants are prepared for subsequent interactions,and a Bridge Feature Extractor (BFE) is specially designed to selectivelyintegrate both high-level and low-level representations to generate thecomprehensive bridge features. Then, instead of conducting heavy pair-wisecross-task interactions, a Task-Feature Refiner (TFR) is developed toefficiently take guidance from bridge features and form final task predictions.To the best of our knowledge, this is the first work considering thecompleteness and quality of feature participants in cross-task interactions.Extensive experiments are conducted on NYUD-v2, Cityscapes and PASCAL Contextbenchmarks, and the superior performance shows the proposed architecture iseffective and powerful in promoting different dense prediction taskssimultaneously.

BridgeNet: Comprehensive and Effective Feature Interactions via Bridge Feature for Multi-task Dense Predictions