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

Automatic crowd behaviour analysis is an important task for intelligenttransportation systems to enable effective flow control and dynamic routeplanning for varying road participants. Crowd counting is one of the keys toautomatic crowd behaviour analysis. Crowd counting using deep convolutionalneural networks (CNN) has achieved encouraging progress in recent years.Researchers have devoted much effort to the design of variant CNN architecturesand most of them are based on the pre-trained VGG16 model. Due to theinsufficient expressive capacity, the backbone network of VGG16 is usuallyfollowed by another cumbersome network specially designed for good countingperformance. Although VGG models have been outperformed by Inception models inimage classification tasks, the existing crowd counting networks built withInception modules still only have a small number of layers with basic types ofInception modules. To fill in this gap, in this paper, we firstly benchmark thebaseline Inception-v3 model on commonly used crowd counting datasets andachieve surprisingly good performance comparable with or better than mostexisting crowd counting models. Subsequently, we push the boundary of thisdisruptive work further by proposing a Segmentation Guided Attention Network(SGANet) with Inception-v3 as the backbone and a novel curriculum loss forcrowd counting. We conduct thorough experiments to compare the performance ofour SGANet with prior arts and the proposed model can achieve state-of-the-artperformance with MAE of 57.6, 6.3 and 87.6 on ShanghaiTechA, ShanghaiTechB andUCF\_QNRF, respectively.

Crowd Counting via Segmentation Guided Attention Networks and Curriculum Loss