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

Fully convolutional neural network (FCN) has been dominating the game of facedetection task for a few years with its congenital capability ofsliding-window-searching with shared kernels, which boiled down all theredundant calculation, and most recent state-of-the-art methods such asFaster-RCNN, SSD, YOLO and FPN use FCN as their backbone. So here comes onequestion: Can we find a universal strategy to further accelerate FCN withhigher accuracy, so could accelerate all the recent FCN-based methods? Toanalyze this, we decompose the face searching space into two orthogonaldirections, `scale' and `spatial'. Only a few coordinates in the space expandedby the two base vectors indicate foreground. So if FCN could ignore most of theother points, the searching space and false alarm should be significantlyboiled down. Based on this philosophy, a novel method named scale estimationand spatial attention proposal ($S^2AP$) is proposed to pay attention to somespecific scales and valid locations in the image pyramid. Furthermore, we adopta masked-convolution operation based on the attention result to accelerate FCNcalculation. Experiments show that FCN-based method RPN can be accelerated byabout $4\times$ with the help of $S^2AP$ and masked-FCN and at the same time itcan also achieve the state-of-the-art on FDDB, AFW and MALF face detectionbenchmarks as well.

Beyond Trade-off: Accelerate FCN-based Face Detector with Higher Accuracy