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

Since convolutional neural network (CNN) lacks an inherent mechanism tohandle large scale variations, we always need to compute feature maps multipletimes for multi-scale object detection, which has the bottleneck ofcomputational cost in practice. To address this, we devise a recurrent scaleapproximation (RSA) to compute feature map once only, and only through this mapcan we approximate the rest maps on other levels. At the core of RSA is therecursive rolling out mechanism: given an initial map at a particular scale, itgenerates the prediction at a smaller scale that is half the size of input. Tofurther increase efficiency and accuracy, we (a): design a scale-forecastnetwork to globally predict potential scales in the image since there is noneed to compute maps on all levels of the pyramid. (b): propose a landmarkretracing network (LRN) to trace back locations of the regressed landmarks andgenerate a confidence score for each landmark; LRN can effectively alleviatefalse positives caused by the accumulated error in RSA. The whole system can betrained end-to-end in a unified CNN framework. Experiments demonstrate that ourproposed algorithm is superior against state-of-the-art methods on facedetection benchmarks and achieves comparable results for generic proposalgeneration. The source code of RSA is available atgithub.com/sciencefans/RSA-for-object-detection.

Recurrent Scale Approximation for Object Detection in CNN