Robust Head Detection in Complex Videos Using Two-Stage Deep Convolution Framework
Pedestrian head detection plays an important role in identifying and localizing individuals in real world visual data. Head detection is a nontrivial problem due to considerable variance in camera view-points, scales, human poses, and appearances in the scene. Thanks to the translation invariance pr...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2020-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9096268/ |
| _version_ | 1818857057247297536 |
|---|---|
| author | Sultan Daud Khan Yasir Ali Basim Zafar Abdulfattah Noorwali |
| author_facet | Sultan Daud Khan Yasir Ali Basim Zafar Abdulfattah Noorwali |
| author_sort | Sultan Daud Khan |
| collection | DOAJ |
| description | Pedestrian head detection plays an important role in identifying and localizing individuals in real world visual data. Head detection is a nontrivial problem due to considerable variance in camera view-points, scales, human poses, and appearances in the scene. Thanks to the translation invariance property of convolutional neural networks (CNNs) which enables large capacity CNNs to handle the problem of appearance and pose variations in the scene. However, the problem of scale invariance is still an open issue. To address this problem, this paper presents a two-stage head detection framework that utilizes fully convolutional network (FCN) to generate scale-aware proposals followed by CNN that classifies each proposal into two classes, i.e. head and background. Experiments results show that using scale-aware proposals obtained by FCN, the object recall rate and mean average precision (mAP) are improved. Additionaly, we demonstrate that our framework achieved state-of-the-art results on four challenging benchmark datasets, i.e. HollywoodHeads, Casablanca, SHOCK, and WIDERFACE. |
| first_indexed | 2024-12-19T08:34:20Z |
| format | Article |
| id | doaj.art-f5cf6823428f4ab8b0b05f276eafedca |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-19T08:34:20Z |
| publishDate | 2020-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-f5cf6823428f4ab8b0b05f276eafedca2022-12-21T20:29:06ZengIEEEIEEE Access2169-35362020-01-018986799869210.1109/ACCESS.2020.29957649096268Robust Head Detection in Complex Videos Using Two-Stage Deep Convolution FrameworkSultan Daud Khan0https://orcid.org/0000-0002-7406-8441Yasir Ali1https://orcid.org/0000-0001-8163-8943Basim Zafar2https://orcid.org/0000-0001-5407-7941Abdulfattah Noorwali3https://orcid.org/0000-0001-9942-2526Department of Computer Science, National University of Technology, Islamabad, PakistanExpert Vision Consulting, Makkah, Saudi ArabiaExpert Vision Consulting, Makkah, Saudi ArabiaDepartment of Electrical Engineering, Umm Al-Qura University, Makkah, Saudi ArabiaPedestrian head detection plays an important role in identifying and localizing individuals in real world visual data. Head detection is a nontrivial problem due to considerable variance in camera view-points, scales, human poses, and appearances in the scene. Thanks to the translation invariance property of convolutional neural networks (CNNs) which enables large capacity CNNs to handle the problem of appearance and pose variations in the scene. However, the problem of scale invariance is still an open issue. To address this problem, this paper presents a two-stage head detection framework that utilizes fully convolutional network (FCN) to generate scale-aware proposals followed by CNN that classifies each proposal into two classes, i.e. head and background. Experiments results show that using scale-aware proposals obtained by FCN, the object recall rate and mean average precision (mAP) are improved. Additionaly, we demonstrate that our framework achieved state-of-the-art results on four challenging benchmark datasets, i.e. HollywoodHeads, Casablanca, SHOCK, and WIDERFACE.https://ieeexplore.ieee.org/document/9096268/Convolutional neural networksnon-maximal suppressionhead detectioncrowd countingmotion analysis |
| spellingShingle | Sultan Daud Khan Yasir Ali Basim Zafar Abdulfattah Noorwali Robust Head Detection in Complex Videos Using Two-Stage Deep Convolution Framework IEEE Access Convolutional neural networks non-maximal suppression head detection crowd counting motion analysis |
| title | Robust Head Detection in Complex Videos Using Two-Stage Deep Convolution Framework |
| title_full | Robust Head Detection in Complex Videos Using Two-Stage Deep Convolution Framework |
| title_fullStr | Robust Head Detection in Complex Videos Using Two-Stage Deep Convolution Framework |
| title_full_unstemmed | Robust Head Detection in Complex Videos Using Two-Stage Deep Convolution Framework |
| title_short | Robust Head Detection in Complex Videos Using Two-Stage Deep Convolution Framework |
| title_sort | robust head detection in complex videos using two stage deep convolution framework |
| topic | Convolutional neural networks non-maximal suppression head detection crowd counting motion analysis |
| url | https://ieeexplore.ieee.org/document/9096268/ |
| work_keys_str_mv | AT sultandaudkhan robustheaddetectionincomplexvideosusingtwostagedeepconvolutionframework AT yasirali robustheaddetectionincomplexvideosusingtwostagedeepconvolutionframework AT basimzafar robustheaddetectionincomplexvideosusingtwostagedeepconvolutionframework AT abdulfattahnoorwali robustheaddetectionincomplexvideosusingtwostagedeepconvolutionframework |