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

Deep learning has shown impressive performance in semantic segmentation, butit is still unaffordable for resource-constrained mobile devices. Whileoffloading computation tasks is promising, the high traffic demands overwhelmthe limited bandwidth. Existing compression algorithms are not fit for semanticsegmentation, as the lack of obvious and concentrated regions of interest(RoIs) forces the adoption of uniform compression strategies, leading to lowcompression ratios or accuracy. This paper introduces STAC, a DNN-drivencompression scheme tailored for edge-assisted semantic video segmentation. STACis the first to exploit DNN's gradients as spatial sensitivity metrics forspatial adaptive compression and achieves superior compression ratio andaccuracy. Yet, it is challenging to adapt this content-customized compressionto videos. Practical issues include varying spatial sensitivity and hugebandwidth consumption for compression strategy feedback and offloading. Wetackle these issues through a spatiotemporal adaptive scheme, which (1) takespartial strategy generation operations offline to reduce communication load,and (2) propagates compression strategies and segmentation results acrossframes through dense optical flow, and adaptively offloads keyframes toaccommodate video content. We implement STAC on a commodity mobile device.Experiments show that STAC can save up to 20.95% of bandwidth without losingaccuracy, compared to the state-of-the-art algorithm.

DNN-Driven Compressive Offloading for Edge-Assisted Semantic Video Segmentation