Accurate Hand Detection from Single-Color Images by Reconstructing Hand Appearances
Hand detection is a crucial pre-processing procedure for many human hand related computer vision tasks, such as hand pose estimation, hand gesture recognition, human activity analysis, and so on. However, reliably detecting multiple hands from cluttering scenes remains to be a challenging task becau...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2019-12-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/20/1/192 |
| _version_ | 1811186449523933184 |
|---|---|
| author | Chi Xu Wendi Cai Yongbo Li Jun Zhou Longsheng Wei |
| author_facet | Chi Xu Wendi Cai Yongbo Li Jun Zhou Longsheng Wei |
| author_sort | Chi Xu |
| collection | DOAJ |
| description | Hand detection is a crucial pre-processing procedure for many human hand related computer vision tasks, such as hand pose estimation, hand gesture recognition, human activity analysis, and so on. However, reliably detecting multiple hands from cluttering scenes remains to be a challenging task because of complex appearance diversities of dexterous human hands (e.g., different hand shapes, skin colors, illuminations, orientations, and scales, etc.) in color images. To tackle this problem, an accurate hand detection method is proposed to reliably detect multiple hands from a single color image using a hybrid detection/reconstruction convolutional neural networks (CNN) framework, in which regions of hands are detected and appearances of hands are reconstructed in parallel by sharing features extracted from a region proposal layer, and the proposed model is trained in an end-to-end manner. Furthermore, it is observed that the generative adversarial network (GAN) could further boost the detection performance by generating more realistic hand appearances. The experimental results show that the proposed approach outperforms the state-of-the-art on public challenging hand detection benchmarks. |
| first_indexed | 2024-04-11T13:46:37Z |
| format | Article |
| id | doaj.art-cbb5c6adbd5c414b97fb22d114dee42a |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-04-11T13:46:37Z |
| publishDate | 2019-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-cbb5c6adbd5c414b97fb22d114dee42a2022-12-22T04:21:03ZengMDPI AGSensors1424-82202019-12-0120119210.3390/s20010192s20010192Accurate Hand Detection from Single-Color Images by Reconstructing Hand AppearancesChi Xu0Wendi Cai1Yongbo Li2Jun Zhou3Longsheng Wei4School of Automation, China University of Geosciences, Wuhan 430074, ChinaSchool of Automation, China University of Geosciences, Wuhan 430074, ChinaSchool of Automation, China University of Geosciences, Wuhan 430074, ChinaSchool of Automation, China University of Geosciences, Wuhan 430074, ChinaSchool of Automation, China University of Geosciences, Wuhan 430074, ChinaHand detection is a crucial pre-processing procedure for many human hand related computer vision tasks, such as hand pose estimation, hand gesture recognition, human activity analysis, and so on. However, reliably detecting multiple hands from cluttering scenes remains to be a challenging task because of complex appearance diversities of dexterous human hands (e.g., different hand shapes, skin colors, illuminations, orientations, and scales, etc.) in color images. To tackle this problem, an accurate hand detection method is proposed to reliably detect multiple hands from a single color image using a hybrid detection/reconstruction convolutional neural networks (CNN) framework, in which regions of hands are detected and appearances of hands are reconstructed in parallel by sharing features extracted from a region proposal layer, and the proposed model is trained in an end-to-end manner. Furthermore, it is observed that the generative adversarial network (GAN) could further boost the detection performance by generating more realistic hand appearances. The experimental results show that the proposed approach outperforms the state-of-the-art on public challenging hand detection benchmarks.https://www.mdpi.com/1424-8220/20/1/192hands detectionhand appearance reconstructionconvolutional neural networksgenerative adversarial networkhuman–computer interaction |
| spellingShingle | Chi Xu Wendi Cai Yongbo Li Jun Zhou Longsheng Wei Accurate Hand Detection from Single-Color Images by Reconstructing Hand Appearances Sensors hands detection hand appearance reconstruction convolutional neural networks generative adversarial network human–computer interaction |
| title | Accurate Hand Detection from Single-Color Images by Reconstructing Hand Appearances |
| title_full | Accurate Hand Detection from Single-Color Images by Reconstructing Hand Appearances |
| title_fullStr | Accurate Hand Detection from Single-Color Images by Reconstructing Hand Appearances |
| title_full_unstemmed | Accurate Hand Detection from Single-Color Images by Reconstructing Hand Appearances |
| title_short | Accurate Hand Detection from Single-Color Images by Reconstructing Hand Appearances |
| title_sort | accurate hand detection from single color images by reconstructing hand appearances |
| topic | hands detection hand appearance reconstruction convolutional neural networks generative adversarial network human–computer interaction |
| url | https://www.mdpi.com/1424-8220/20/1/192 |
| work_keys_str_mv | AT chixu accuratehanddetectionfromsinglecolorimagesbyreconstructinghandappearances AT wendicai accuratehanddetectionfromsinglecolorimagesbyreconstructinghandappearances AT yongboli accuratehanddetectionfromsinglecolorimagesbyreconstructinghandappearances AT junzhou accuratehanddetectionfromsinglecolorimagesbyreconstructinghandappearances AT longshengwei accuratehanddetectionfromsinglecolorimagesbyreconstructinghandappearances |