Relatable Clothing: Soft-Attention Mechanism for Detecting Worn/Unworn Objects
We have identified a need in visual relationship detection and biometrics related research for a dataset and model which focuses on person-clothing pairs. Previous to our Relatable Clothing dataset, there were no publicly available datasets usable for “worn” and “unw...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2021-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9503396/ |
| _version_ | 1818576127028887552 |
|---|---|
| author | Thomas Truong Svetlana Yanushkevich |
| author_facet | Thomas Truong Svetlana Yanushkevich |
| author_sort | Thomas Truong |
| collection | DOAJ |
| description | We have identified a need in visual relationship detection and biometrics related research for a dataset and model which focuses on person-clothing pairs. Previous to our Relatable Clothing dataset, there were no publicly available datasets usable for “worn” and “unworn” clothing detection. In this paper we propose a novel visual relationship model architecture for “worn” and “unworn” clothing detection that makes use of a soft attention mechanism for feature fusion between a conventional ResNet backbone and our novel person-clothing mask feature extraction architecture. The best proposed model achieves 98.62% accuracy, 99.50% precision, 98.31% recall, and 99.14% specificity on the Relatable Clothing dataset, outperforming our previous iterations. We release our models which can be found on the Relatable Clothing GitHub repository (<uri>https://github.com/th-truong/relatable_clothing</uri>) for future research and applications into detecting and analyzing person-clothing pairs. |
| first_indexed | 2024-12-15T00:49:58Z |
| format | Article |
| id | doaj.art-8ef30961605d46739d93991aab5a0ec2 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-15T00:49:58Z |
| publishDate | 2021-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-8ef30961605d46739d93991aab5a0ec22022-12-21T22:41:27ZengIEEEIEEE Access2169-35362021-01-01910878210879210.1109/ACCESS.2021.31017899503396Relatable Clothing: Soft-Attention Mechanism for Detecting Worn/Unworn ObjectsThomas Truong0https://orcid.org/0000-0002-9934-4553Svetlana Yanushkevich1https://orcid.org/0000-0003-4794-9849Department of Electrical and Software Engineering, Biometric Technologies Laboratory, University of Calgary, Calgary, AB, CanadaDepartment of Electrical and Software Engineering, Biometric Technologies Laboratory, University of Calgary, Calgary, AB, CanadaWe have identified a need in visual relationship detection and biometrics related research for a dataset and model which focuses on person-clothing pairs. Previous to our Relatable Clothing dataset, there were no publicly available datasets usable for “worn” and “unworn” clothing detection. In this paper we propose a novel visual relationship model architecture for “worn” and “unworn” clothing detection that makes use of a soft attention mechanism for feature fusion between a conventional ResNet backbone and our novel person-clothing mask feature extraction architecture. The best proposed model achieves 98.62% accuracy, 99.50% precision, 98.31% recall, and 99.14% specificity on the Relatable Clothing dataset, outperforming our previous iterations. We release our models which can be found on the Relatable Clothing GitHub repository (<uri>https://github.com/th-truong/relatable_clothing</uri>) for future research and applications into detecting and analyzing person-clothing pairs.https://ieeexplore.ieee.org/document/9503396/Visual relationship detectioncomputer visionsoft biometricsobject detectionneural networks |
| spellingShingle | Thomas Truong Svetlana Yanushkevich Relatable Clothing: Soft-Attention Mechanism for Detecting Worn/Unworn Objects IEEE Access Visual relationship detection computer vision soft biometrics object detection neural networks |
| title | Relatable Clothing: Soft-Attention Mechanism for Detecting Worn/Unworn Objects |
| title_full | Relatable Clothing: Soft-Attention Mechanism for Detecting Worn/Unworn Objects |
| title_fullStr | Relatable Clothing: Soft-Attention Mechanism for Detecting Worn/Unworn Objects |
| title_full_unstemmed | Relatable Clothing: Soft-Attention Mechanism for Detecting Worn/Unworn Objects |
| title_short | Relatable Clothing: Soft-Attention Mechanism for Detecting Worn/Unworn Objects |
| title_sort | relatable clothing soft attention mechanism for detecting worn unworn objects |
| topic | Visual relationship detection computer vision soft biometrics object detection neural networks |
| url | https://ieeexplore.ieee.org/document/9503396/ |
| work_keys_str_mv | AT thomastruong relatableclothingsoftattentionmechanismfordetectingwornunwornobjects AT svetlanayanushkevich relatableclothingsoftattentionmechanismfordetectingwornunwornobjects |