Presentation 2021-12-16
Evaluation of CNNs with Narrow Receptive Fields in Crowd Counting
Junya Uchida, Yu Wang, Jien Kato,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Density map estimation for crowd counting means to estimate the number of people by regressing a two-dimensional density map that contains the quantitative and positional information of people’s head. To appropriately deal with images with different densities and scales of people, it is important to investigate how the receptive field size, a range that contributes to the recognition accuracy, affects the model performance. In this paper, we especially evaluate the effect of narrow receptive fields on counting accuracy. Furthermore, on the basis of our evaluation, we propose a model that achieves better accuracy by fusing the networks with broad and narrow receptive fields.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Crowd Counting / CNNs / Receptive Fields / BagNet
Paper # PRMU2021-43
Date of Issue 2021-12-09 (PRMU)

Conference Information
Committee PRMU
Conference Date 2021/12/16(2days)
Place (in Japanese) (See Japanese page)
Place (in English) Online
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair Seiichi Uchida(Kyushu Univ.)
Vice Chair Masakazu Iwamura(Osaka Pref. Univ.) / Mitsuru Anpai(Denso IT Lab.)
Secretary Masakazu Iwamura(NTT) / Mitsuru Anpai(Tottori Univ.)
Assistant Kouta Yamaguchi(CyberAgent) / Yusuke Matsui(Univ. of Tokyo)

Paper Information
Registration To Technical Committee on Pattern Recognition and Media Understanding
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Evaluation of CNNs with Narrow Receptive Fields in Crowd Counting
Sub Title (in English)
Keyword(1) Crowd Counting
Keyword(2) CNNs
Keyword(3) Receptive Fields
Keyword(4) BagNet
1st Author's Name Junya Uchida
1st Author's Affiliation Ritsumeikan University(Ritsumeikan Univ.)
2nd Author's Name Yu Wang
2nd Author's Affiliation Ritsumeikan University(Ritsumeikan Univ.)
3rd Author's Name Jien Kato
3rd Author's Affiliation Ritsumeikan University(Ritsumeikan Univ.)
Date 2021-12-16
Paper # PRMU2021-43
Volume (vol) vol.121
Number (no) PRMU-304
Page pp.pp.101-106(PRMU),
#Pages 6
Date of Issue 2021-12-09 (PRMU)