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

Multi-modality image fusion and segmentation play a vital role in autonomousdriving and robotic operation. Early efforts focus on boosting the performancefor only one task, \emph{e.g.,} fusion or segmentation, making it hard toreach~`Best of Both Worlds'. To overcome this issue, in this paper, we proposea \textbf{M}ulti-\textbf{i}nteractive \textbf{F}eature learning architecturefor image fusion and \textbf{Seg}mentation, namely SegMiF, and exploitdual-task correlation to promote the performance of both tasks. The SegMiF isof a cascade structure, containing a fusion sub-network and a commonly usedsegmentation sub-network. By slickly bridging intermediate features between twocomponents, the knowledge learned from the segmentation task can effectivelyassist the fusion task. Also, the benefited fusion network supports thesegmentation one to perform more pretentiously. Besides, a hierarchicalinteractive attention block is established to ensure fine-grained mapping ofall the vital information between two tasks, so that the modality/semanticfeatures can be fully mutual-interactive. In addition, a dynamic weight factoris introduced to automatically adjust the corresponding weights of each task,which can balance the interactive feature correspondence and break through thelimitation of laborious tuning. Furthermore, we construct a smart multi-wavebinocular imaging system and collect a full-time multi-modality benchmark with15 annotated pixel-level categories for image fusion and segmentation.Extensive experiments on several public datasets and our benchmark demonstratethat the proposed method outputs visually appealing fused images and performaveragely $7.66\%$ higher segmentation mIoU in the real-world scene than thestate-of-the-art approaches. The source code and benchmark are available at\url{https://github.com/JinyuanLiu-CV/SegMiF}.

Multi-interactive Feature Learning and a Full-time Multi-modality Benchmark for Image Fusion and Segmentation