Improving the performance of pedestrian detectors using convolutional learning
详细信息 查看全文
- 作者：David Ribeiroa ; david.ribeiro@ist.utl.pt" class="auth_mail" title="E-mail the corresponding authorAuthor Vitae ; Jacinto C. Nascimentoa ; jan@isr.ist.utl.pt" class="auth_mail" title="E-mail the corresponding authorAuthor Vitae ; Alexandre Bernardinoa ; alex@isr.ist.utl.pt" class="auth_mail" title="E-mail the corresponding authorAuthor Vitae ; Gustavo Carneirob ; gcarneiro@gmail.com" class="auth_mail" title="E-mail the corresponding authorAuthor Vitae
- 关键词：Pedestrian detection ; Convolutional neural network ; Feature maps ; Non-deep detectors
- 刊名：Pattern Recognition
- 出版年：2017
- 出版时间：January 2017
- 年：2017
- 卷：61
- 期：Complete
- 页码：641-649
- 全文大小：3422 K
文摘
- class="listitem" id="list_li0005">
- class="label">•
This paper proposes a cascade of non-deep detectors with a CNN model.
- class="label">•
Pedestrian Detection is addressed by testing on INRIA and Caltech datasets.
- class="label">•
Different non-deep detectors are used: ACF, LDCF and Spatial Pooling + (SP+).
- class="label">•
State-of-the-art competitive performance is achieved cascading SP+ with the CNN.