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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Real-time semantic segmentation is desirable in many robotic applicationswith limited computation resources. One challenge of semantic segmentation isto deal with the object scale variations and leverage the context. How toperform multi-scale context aggregation within limited computation budget isimportant. In this paper, firstly, we introduce a novel and efficient modulecalled Cascaded Factorized Atrous Spatial Pyramid Pooling (CF-ASPP). It is alightweight cascaded structure for Convolutional Neural Networks (CNNs) toefficiently leverage context information. On the other hand, for runtimeefficiency, state-of-the-art methods will quickly decrease the spatial size ofthe inputs or feature maps in the early network stages. The finalhigh-resolution result is usually obtained by non-parametric up-samplingoperation (e.g. bilinear interpolation). Differently, we rethink this pipelineand treat it as a super-resolution process. We use optimized super-resolutionoperation in the up-sampling step and improve the accuracy, especially insub-sampled input image scenario for real-time applications. By fusing theabove two improvements, our methods provide better latency-accuracy trade-offthan the other state-of-the-art methods. In particular, we achieve 68.4% mIoUat 84 fps on the Cityscapes test set with a single Nivida Titan X (Maxwell) GPUcard. The proposed module can be plugged into any feature extraction CNN andbenefits from the CNN structure development.

FarSee-Net: Real-Time Semantic Segmentation by Efficient Multi-scale Context Aggregation and Feature Space Super-resolution