File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Article: DeepID-Net: Object Detection with Deformable Part Based Convolutional Neural Networks

TitleDeepID-Net: Object Detection with Deformable Part Based Convolutional Neural Networks
Authors
Keywordsdeep learning
convolutional neural networks
CNN
deep model
object detection
Issue Date2017
Citation
IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, v. 39, n. 7, p. 1320-1334 How to Cite?
Abstract© 2017 IEEE. In this paper, we propose deformable deep convolutional neural networks for generic object detection. This new deep learning object detection framework has innovations in multiple aspects. In the proposed new deep architecture, a new deformation constrained pooling (def-pooling) layer models the deformation of object parts with geometric constraint and penalty. A new pre-training strategy is proposed to learn feature representations more suitable for the object detection task and with good generalization capability. By changing the net structures, training strategies, adding and removing some key components in the detection pipeline, a set of models with large diversity are obtained, which significantly improves the effectiveness of model averaging. The proposed approach improves the mean averaged precision obtained by RCNN [1], which was the state-of-the-art, from 31 to 50.3 percent on the ILSVRC2014 detection test set. It also outperforms the winner of ILSVRC2014, GoogLeNet, by 6.1 percent. Detailed component-wise analysis is also provided through extensive experimental evaluation, which provides a global view for people to understand the deep learning object detection pipeline.
Persistent Identifierhttp://hdl.handle.net/10722/273598
ISSN
2023 Impact Factor: 20.8
2023 SCImago Journal Rankings: 6.158
ISI Accession Number ID

 

DC FieldValueLanguage
dc.contributor.authorOuyang, Wanli-
dc.contributor.authorZeng, Xingyu-
dc.contributor.authorWang, Xiaogang-
dc.contributor.authorQiu, Shi-
dc.contributor.authorLuo, Ping-
dc.contributor.authorTian, Yonglong-
dc.contributor.authorLi, Hongsheng-
dc.contributor.authorYang, Shuo-
dc.contributor.authorWang, Zhe-
dc.contributor.authorLi, Hongyang-
dc.contributor.authorWang, Kun-
dc.contributor.authorYan, Junjie-
dc.contributor.authorLoy, Chen Change-
dc.contributor.authorTang, Xiaoou-
dc.date.accessioned2019-08-12T09:56:05Z-
dc.date.available2019-08-12T09:56:05Z-
dc.date.issued2017-
dc.identifier.citationIEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, v. 39, n. 7, p. 1320-1334-
dc.identifier.issn0162-8828-
dc.identifier.urihttp://hdl.handle.net/10722/273598-
dc.description.abstract© 2017 IEEE. In this paper, we propose deformable deep convolutional neural networks for generic object detection. This new deep learning object detection framework has innovations in multiple aspects. In the proposed new deep architecture, a new deformation constrained pooling (def-pooling) layer models the deformation of object parts with geometric constraint and penalty. A new pre-training strategy is proposed to learn feature representations more suitable for the object detection task and with good generalization capability. By changing the net structures, training strategies, adding and removing some key components in the detection pipeline, a set of models with large diversity are obtained, which significantly improves the effectiveness of model averaging. The proposed approach improves the mean averaged precision obtained by RCNN [1], which was the state-of-the-art, from 31 to 50.3 percent on the ILSVRC2014 detection test set. It also outperforms the winner of ILSVRC2014, GoogLeNet, by 6.1 percent. Detailed component-wise analysis is also provided through extensive experimental evaluation, which provides a global view for people to understand the deep learning object detection pipeline.-
dc.languageeng-
dc.relation.ispartofIEEE Transactions on Pattern Analysis and Machine Intelligence-
dc.subjectdeep learning-
dc.subjectconvolutional neural networks-
dc.subjectCNN-
dc.subjectdeep model-
dc.subjectobject detection-
dc.titleDeepID-Net: Object Detection with Deformable Part Based Convolutional Neural Networks-
dc.typeArticle-
dc.description.naturelink_to_subscribed_fulltext-
dc.identifier.doi10.1109/TPAMI.2016.2587642-
dc.identifier.pmid27392342-
dc.identifier.scopuseid_2-s2.0-85020430340-
dc.identifier.volume39-
dc.identifier.issue7-
dc.identifier.spage1320-
dc.identifier.epage1334-
dc.identifier.isiWOS:000402744400004-
dc.identifier.issnl0162-8828-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats