Transport Object Detection in Street View Imagery Using Decomposed Convolutional Neural Networks
Deep learning has achieved great successes in performing many visual recognition tasks, including object detection. Nevertheless, existing deep networks are computationally expensive and memory intensive, hindering their deployment in resource-constrained environments, such as mobile or embedded dev...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-09-01
|
| Series: | Mathematics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2227-7390/11/18/3839 |
| _version_ | 1797578954709663744 |
|---|---|
| author | Yunpeng Bai Changjing Shang Ying Li Liang Shen Shangzhu Jin Qiang Shen |
| author_facet | Yunpeng Bai Changjing Shang Ying Li Liang Shen Shangzhu Jin Qiang Shen |
| author_sort | Yunpeng Bai |
| collection | DOAJ |
| description | Deep learning has achieved great successes in performing many visual recognition tasks, including object detection. Nevertheless, existing deep networks are computationally expensive and memory intensive, hindering their deployment in resource-constrained environments, such as mobile or embedded devices that are widely used by city travellers. Recently, estimating city-level travel patterns using street imagery has been shown to be a potentially valid way according to a case study with Google Street View (GSV), addressing a critical challenge in transport object detection. This paper presents a compressed deep network using tensor decomposition to detect transport objects in GSV images, which is sustainable and eco-friendly. In particular, a new dataset named Transport Mode Share-Tokyo (TMS-Tokyo) is created to serve the public for transport object detection. This is based on the selection and filtering of 32,555 acquired images that involve 50,827 visible transport objects (including cars, pedestrians, buses, trucks, motors, vans, cyclists and parked bicycles) from the GSV imagery of Tokyo. Then a compressed convolutional neural network (termed SVDet) is proposed for street view object detection via tensor train decomposition on a given baseline detector. The method proposed herein yields a mean average precision (mAP) of 77.6% on the newly introduced dataset, TMS-Tokyo, necessitating just 17.29 M parameters and a computational capacity of 16.52 G FLOPs. As such, it markedly surpasses the performance of existing state-of-the-art methods documented in the literature. |
| first_indexed | 2024-03-10T22:29:58Z |
| format | Article |
| id | doaj.art-6379262c59e2472e9b6539ff72984be2 |
| institution | Directory Open Access Journal |
| issn | 2227-7390 |
| language | English |
| last_indexed | 2024-03-10T22:29:58Z |
| publishDate | 2023-09-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Mathematics |
| spelling | doaj.art-6379262c59e2472e9b6539ff72984be22023-11-19T11:48:18ZengMDPI AGMathematics2227-73902023-09-011118383910.3390/math11183839Transport Object Detection in Street View Imagery Using Decomposed Convolutional Neural NetworksYunpeng Bai0Changjing Shang1Ying Li2Liang Shen3Shangzhu Jin4Qiang Shen5Department of Computer Science, Aberystwyth University, Aberystwyth SY23 3DB, UKDepartment of Computer Science, Aberystwyth University, Aberystwyth SY23 3DB, UKSchool of Computer Science, Northwestern Polytechnical University, Xi’an 710129, ChinaSchool of Information Engineering, Fujian Business University, Fuzhou 350506, ChinaInformation Office, Chongqing University of Science and Technology, Chongqing 401331, ChinaSchool of Computer Science, Northwestern Polytechnical University, Xi’an 710129, ChinaDeep learning has achieved great successes in performing many visual recognition tasks, including object detection. Nevertheless, existing deep networks are computationally expensive and memory intensive, hindering their deployment in resource-constrained environments, such as mobile or embedded devices that are widely used by city travellers. Recently, estimating city-level travel patterns using street imagery has been shown to be a potentially valid way according to a case study with Google Street View (GSV), addressing a critical challenge in transport object detection. This paper presents a compressed deep network using tensor decomposition to detect transport objects in GSV images, which is sustainable and eco-friendly. In particular, a new dataset named Transport Mode Share-Tokyo (TMS-Tokyo) is created to serve the public for transport object detection. This is based on the selection and filtering of 32,555 acquired images that involve 50,827 visible transport objects (including cars, pedestrians, buses, trucks, motors, vans, cyclists and parked bicycles) from the GSV imagery of Tokyo. Then a compressed convolutional neural network (termed SVDet) is proposed for street view object detection via tensor train decomposition on a given baseline detector. The method proposed herein yields a mean average precision (mAP) of 77.6% on the newly introduced dataset, TMS-Tokyo, necessitating just 17.29 M parameters and a computational capacity of 16.52 G FLOPs. As such, it markedly surpasses the performance of existing state-of-the-art methods documented in the literature.https://www.mdpi.com/2227-7390/11/18/3839convolutional neural networksstreet-view object detectiontensor train decomposition |
| spellingShingle | Yunpeng Bai Changjing Shang Ying Li Liang Shen Shangzhu Jin Qiang Shen Transport Object Detection in Street View Imagery Using Decomposed Convolutional Neural Networks Mathematics convolutional neural networks street-view object detection tensor train decomposition |
| title | Transport Object Detection in Street View Imagery Using Decomposed Convolutional Neural Networks |
| title_full | Transport Object Detection in Street View Imagery Using Decomposed Convolutional Neural Networks |
| title_fullStr | Transport Object Detection in Street View Imagery Using Decomposed Convolutional Neural Networks |
| title_full_unstemmed | Transport Object Detection in Street View Imagery Using Decomposed Convolutional Neural Networks |
| title_short | Transport Object Detection in Street View Imagery Using Decomposed Convolutional Neural Networks |
| title_sort | transport object detection in street view imagery using decomposed convolutional neural networks |
| topic | convolutional neural networks street-view object detection tensor train decomposition |
| url | https://www.mdpi.com/2227-7390/11/18/3839 |
| work_keys_str_mv | AT yunpengbai transportobjectdetectioninstreetviewimageryusingdecomposedconvolutionalneuralnetworks AT changjingshang transportobjectdetectioninstreetviewimageryusingdecomposedconvolutionalneuralnetworks AT yingli transportobjectdetectioninstreetviewimageryusingdecomposedconvolutionalneuralnetworks AT liangshen transportobjectdetectioninstreetviewimageryusingdecomposedconvolutionalneuralnetworks AT shangzhujin transportobjectdetectioninstreetviewimageryusingdecomposedconvolutionalneuralnetworks AT qiangshen transportobjectdetectioninstreetviewimageryusingdecomposedconvolutionalneuralnetworks |