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

Recently, many methods have been proposed for object detection. They cannotdetect objects by semantic features, adaptively. In this work, according tochannel and spatial attention mechanisms, we mainly analyze that differentmethods detect objects adaptively. Some state-of-the-art detectors combinedifferent feature pyramids with many mechanisms to enhance multi-level semanticinformation. However, they require more cost. This work addresses that by ananchor-free detector with shared encoder-decoder with attention mechanism,extracting shared features. We consider features of different levels frombackbone (e.g., ResNet-50) as the basis features. Then, we feed the featuresinto a simple module, followed by a detector header to detect objects.Meantime, we use the semantic features to revise geometric locations, and thedetector is a pixel-semantic revising of position. More importantly, this workanalyzes the impact of different pooling strategies (e.g., mean, maximum orminimum) on multi-scale objects, and finds the minimum pooling improvedetection performance on small objects better. Compared with state-of-the-artMNC based on ResNet-101 for the standard MSCOCO 2014 baseline, our methodimproves detection AP of 3.8%.

Pixel-Semantic Revise of Position Learning A One-Stage Object Detector with A Shared Encoder-Decoder