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-modal fusion is increasingly being used for autonomous driving tasks,as images from different modalities provide unique information for featureextraction. However, the existing two-stream networks are only fused at aspecific network layer, which requires a lot of manual attempts to set up. Asthe CNN goes deeper, the two modal features become more and more advanced andabstract, and the fusion occurs at the feature level with a large gap, whichcan easily hurt the performance. In this study, we propose a novel fusionarchitecture called skip-cross networks (SkipcrossNets), which combinesadaptively LiDAR point clouds and camera images without being bound to acertain fusion epoch. Specifically, skip-cross connects each layer to eachlayer in a feed-forward manner, and for each layer, the feature maps of allprevious layers are used as input and its own feature maps are used as input toall subsequent layers for the other modality, enhancing feature propagation andmulti-modal features fusion. This strategy facilitates selection of the mostsimilar feature layers from two data pipelines, providing a complementaryeffect for sparse point cloud features during fusion processes. The network isalso divided into several blocks to reduce the complexity of feature fusion andthe number of model parameters. The advantages of skip-cross fusion weredemonstrated through application to the KITTI and A2D2 datasets, achieving aMaxF score of 96.85% on KITTI and an F1 score of 84.84% on A2D2. The modelparameters required only 2.33 MB of memory at a speed of 68.24 FPS, which couldbe viable for mobile terminals and embedded devices.

SkipcrossNets: Adaptive Skip-cross Fusion for Road Detection