File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Conference Paper: Unconstrained fashion landmark detection via hierarchical recurrent transformer networks

TitleUnconstrained fashion landmark detection via hierarchical recurrent transformer networks
Authors
KeywordsDeep learning
Convolutional neural network
Visual fashion understanding
Landmark detection
Issue Date2017
Citation
MM 2017 - Proceedings of the 2017 ACM Multimedia Conference, 2017, p. 172-180 How to Cite?
Abstract© 2017 Copyright held by the owner/author(s). Fashion landmarks are functional key points defined on clothes, such as corners of neckline, hemline, and cuff. They have been recently introduced [18] as an effective visual representation for fashion image understanding. However, detecting fashion landmarks are challenging due to background clutters, human poses, and scales as shown in Fig. 1. To remove the above variations, previous works usually assumed bounding boxes of clothes are provided in training and test as additional annotations, which are expensive to obtain and inapplicable in practice. This work addresses unconstrained fashion landmark detection, where clothing bounding boxes are not provided in both training and test. To this end, we present a novel Deep LAndmark Network (DLAN), where bounding boxes and landmarks are jointly estimated and trained iteratively in an end-to-end manner. DLAN contains two dedicated modules, including a Selective Dilated Convolution for handling scale discrepancies, and a Hierarchical Recurrent Spatial Transformer for handling background clutters. To evaluate DLAN, we present a large-scale fashion landmark dataset, namely Unconstrained Landmark Database (ULD), consisting of 30K images. Statistics show that ULD is more challenging than existing datasets in terms of image scales, background clutters, and human poses. Extensive experiments demonstrate the effectiveness of DLAN over the state-of-the-art methods. DLAN also exhibits excellent generalization across different clothing categories and modalities, making it extremely suitable for real-world fashion analysis.
Persistent Identifierhttp://hdl.handle.net/10722/273679

 

DC FieldValueLanguage
dc.contributor.authorYan, Sijie-
dc.contributor.authorLiu, Ziwei-
dc.contributor.authorLuo, Ping-
dc.contributor.authorQiu, Shi-
dc.contributor.authorWang, Xiaogang-
dc.contributor.authorTang, Xiaoou-
dc.date.accessioned2019-08-12T09:56:21Z-
dc.date.available2019-08-12T09:56:21Z-
dc.date.issued2017-
dc.identifier.citationMM 2017 - Proceedings of the 2017 ACM Multimedia Conference, 2017, p. 172-180-
dc.identifier.urihttp://hdl.handle.net/10722/273679-
dc.description.abstract© 2017 Copyright held by the owner/author(s). Fashion landmarks are functional key points defined on clothes, such as corners of neckline, hemline, and cuff. They have been recently introduced [18] as an effective visual representation for fashion image understanding. However, detecting fashion landmarks are challenging due to background clutters, human poses, and scales as shown in Fig. 1. To remove the above variations, previous works usually assumed bounding boxes of clothes are provided in training and test as additional annotations, which are expensive to obtain and inapplicable in practice. This work addresses unconstrained fashion landmark detection, where clothing bounding boxes are not provided in both training and test. To this end, we present a novel Deep LAndmark Network (DLAN), where bounding boxes and landmarks are jointly estimated and trained iteratively in an end-to-end manner. DLAN contains two dedicated modules, including a Selective Dilated Convolution for handling scale discrepancies, and a Hierarchical Recurrent Spatial Transformer for handling background clutters. To evaluate DLAN, we present a large-scale fashion landmark dataset, namely Unconstrained Landmark Database (ULD), consisting of 30K images. Statistics show that ULD is more challenging than existing datasets in terms of image scales, background clutters, and human poses. Extensive experiments demonstrate the effectiveness of DLAN over the state-of-the-art methods. DLAN also exhibits excellent generalization across different clothing categories and modalities, making it extremely suitable for real-world fashion analysis.-
dc.languageeng-
dc.relation.ispartofMM 2017 - Proceedings of the 2017 ACM Multimedia Conference-
dc.subjectDeep learning-
dc.subjectConvolutional neural network-
dc.subjectVisual fashion understanding-
dc.subjectLandmark detection-
dc.titleUnconstrained fashion landmark detection via hierarchical recurrent transformer networks-
dc.typeConference_Paper-
dc.description.naturelink_to_subscribed_fulltext-
dc.identifier.doi10.1145/3123266.3123276-
dc.identifier.scopuseid_2-s2.0-85035239756-
dc.identifier.spage172-
dc.identifier.epage180-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats