A New Dataset and Performance Evaluation of a Region-Based CNN for Urban Object Detection
In recent years, we have seen a large growth in the number of applications which use deep learning-based object detectors. Autonomous driving assistance systems (ADAS) are one of the areas where they have the most impact. This work presents a novel study evaluating a state-of-the-art technique for u...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2018-11-01
|
| Series: | Electronics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2079-9292/7/11/301 |
| _version_ | 1811185279646564352 |
|---|---|
| author | Alex Dominguez-Sanchez Miguel Cazorla Sergio Orts-Escolano |
| author_facet | Alex Dominguez-Sanchez Miguel Cazorla Sergio Orts-Escolano |
| author_sort | Alex Dominguez-Sanchez |
| collection | DOAJ |
| description | In recent years, we have seen a large growth in the number of applications which use deep learning-based object detectors. Autonomous driving assistance systems (ADAS) are one of the areas where they have the most impact. This work presents a novel study evaluating a state-of-the-art technique for urban object detection and localization. In particular, we investigated the performance of the Faster R-CNN method to detect and localize urban objects in a variety of outdoor urban videos involving pedestrians, cars, bicycles and other objects moving in the scene (urban driving). We propose a new dataset that is used for benchmarking the accuracy of a real-time object detector (Faster R-CNN). Part of the data was collected using an HD camera mounted on a vehicle. Furthermore, some of the data is weakly annotated so it can be used for testing weakly supervised learning techniques. There already exist urban object datasets, but none of them include all the essential urban objects. We carried out extensive experiments demonstrating the effectiveness of the baseline approach. Additionally, we propose an R-CNN plus tracking technique to accelerate the process of real-time urban object detection. |
| first_indexed | 2024-04-11T13:27:55Z |
| format | Article |
| id | doaj.art-7dd8a0e7150d43229e041c25740ca7a8 |
| institution | Directory Open Access Journal |
| issn | 2079-9292 |
| language | English |
| last_indexed | 2024-04-11T13:27:55Z |
| publishDate | 2018-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Electronics |
| spelling | doaj.art-7dd8a0e7150d43229e041c25740ca7a82022-12-22T04:22:00ZengMDPI AGElectronics2079-92922018-11-0171130110.3390/electronics7110301electronics7110301A New Dataset and Performance Evaluation of a Region-Based CNN for Urban Object DetectionAlex Dominguez-Sanchez0Miguel Cazorla1Sergio Orts-Escolano2RoViT, University of Alicante, Carretera San Vicente del Raspeig s/n 03690, San Vicente del Raspeig, Alicante 03690, SpainRoViT, University of Alicante, Carretera San Vicente del Raspeig s/n 03690, San Vicente del Raspeig, Alicante 03690, SpainRoViT, University of Alicante, Carretera San Vicente del Raspeig s/n 03690, San Vicente del Raspeig, Alicante 03690, SpainIn recent years, we have seen a large growth in the number of applications which use deep learning-based object detectors. Autonomous driving assistance systems (ADAS) are one of the areas where they have the most impact. This work presents a novel study evaluating a state-of-the-art technique for urban object detection and localization. In particular, we investigated the performance of the Faster R-CNN method to detect and localize urban objects in a variety of outdoor urban videos involving pedestrians, cars, bicycles and other objects moving in the scene (urban driving). We propose a new dataset that is used for benchmarking the accuracy of a real-time object detector (Faster R-CNN). Part of the data was collected using an HD camera mounted on a vehicle. Furthermore, some of the data is weakly annotated so it can be used for testing weakly supervised learning techniques. There already exist urban object datasets, but none of them include all the essential urban objects. We carried out extensive experiments demonstrating the effectiveness of the baseline approach. Additionally, we propose an R-CNN plus tracking technique to accelerate the process of real-time urban object detection.https://www.mdpi.com/2079-9292/7/11/301real-time object detectionautonomous driving assistance systemurban object detectorconvolutional neural networks |
| spellingShingle | Alex Dominguez-Sanchez Miguel Cazorla Sergio Orts-Escolano A New Dataset and Performance Evaluation of a Region-Based CNN for Urban Object Detection Electronics real-time object detection autonomous driving assistance system urban object detector convolutional neural networks |
| title | A New Dataset and Performance Evaluation of a Region-Based CNN for Urban Object Detection |
| title_full | A New Dataset and Performance Evaluation of a Region-Based CNN for Urban Object Detection |
| title_fullStr | A New Dataset and Performance Evaluation of a Region-Based CNN for Urban Object Detection |
| title_full_unstemmed | A New Dataset and Performance Evaluation of a Region-Based CNN for Urban Object Detection |
| title_short | A New Dataset and Performance Evaluation of a Region-Based CNN for Urban Object Detection |
| title_sort | new dataset and performance evaluation of a region based cnn for urban object detection |
| topic | real-time object detection autonomous driving assistance system urban object detector convolutional neural networks |
| url | https://www.mdpi.com/2079-9292/7/11/301 |
| work_keys_str_mv | AT alexdominguezsanchez anewdatasetandperformanceevaluationofaregionbasedcnnforurbanobjectdetection AT miguelcazorla anewdatasetandperformanceevaluationofaregionbasedcnnforurbanobjectdetection AT sergioortsescolano anewdatasetandperformanceevaluationofaregionbasedcnnforurbanobjectdetection AT alexdominguezsanchez newdatasetandperformanceevaluationofaregionbasedcnnforurbanobjectdetection AT miguelcazorla newdatasetandperformanceevaluationofaregionbasedcnnforurbanobjectdetection AT sergioortsescolano newdatasetandperformanceevaluationofaregionbasedcnnforurbanobjectdetection |