A Review of Video Object Detection: Datasets, Metrics and Methods
Although there are well established object detection methods based on static images, their application to video data on a frame by frame basis faces two shortcomings: (i) lack of computational efficiency due to redundancy across image frames or by not using a temporal and spatial correlation of feat...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2020-11-01
|
Series: | Applied Sciences |
Subjects: | |
Online Access: | https://www.mdpi.com/2076-3417/10/21/7834 |
_version_ | 1797548834515058688 |
---|---|
author | Haidi Zhu Haoran Wei Baoqing Li Xiaobing Yuan Nasser Kehtarnavaz |
author_facet | Haidi Zhu Haoran Wei Baoqing Li Xiaobing Yuan Nasser Kehtarnavaz |
author_sort | Haidi Zhu |
collection | DOAJ |
description | Although there are well established object detection methods based on static images, their application to video data on a frame by frame basis faces two shortcomings: (i) lack of computational efficiency due to redundancy across image frames or by not using a temporal and spatial correlation of features across image frames, and (ii) lack of robustness to real-world conditions such as motion blur and occlusion. Since the introduction of the ImageNet Large Scale Visual Recognition Challenge (ILSVRC) in 2015, a growing number of methods have appeared in the literature on video object detection, many of which have utilized deep learning models. The aim of this paper is to provide a review of these papers on video object detection. An overview of the existing datasets for video object detection together with commonly used evaluation metrics is first presented. Video object detection methods are then categorized and a description of each of them is stated. Two comparison tables are provided to see their differences in terms of both accuracy and computational efficiency. Finally, some future trends in video object detection to address the challenges involved are noted. |
first_indexed | 2024-03-10T15:05:24Z |
format | Article |
id | doaj.art-b67ff858b2e04b888b34514fe3cf1511 |
institution | Directory Open Access Journal |
issn | 2076-3417 |
language | English |
last_indexed | 2024-03-10T15:05:24Z |
publishDate | 2020-11-01 |
publisher | MDPI AG |
record_format | Article |
series | Applied Sciences |
spelling | doaj.art-b67ff858b2e04b888b34514fe3cf15112023-11-20T19:49:20ZengMDPI AGApplied Sciences2076-34172020-11-011021783410.3390/app10217834A Review of Video Object Detection: Datasets, Metrics and MethodsHaidi Zhu0Haoran Wei1Baoqing Li2Xiaobing Yuan3Nasser Kehtarnavaz4Science and Technology on Micro-System Laboratory, Shanghai Institute of Microsystem and Information Technology, Chinese Academy of Sciences, Shanghai 201800, ChinaDepartment of Electrical and Computer Engineering, University of Texas at Dallas, Richardson, TX 75080, USAScience and Technology on Micro-System Laboratory, Shanghai Institute of Microsystem and Information Technology, Chinese Academy of Sciences, Shanghai 201800, ChinaScience and Technology on Micro-System Laboratory, Shanghai Institute of Microsystem and Information Technology, Chinese Academy of Sciences, Shanghai 201800, ChinaDepartment of Electrical and Computer Engineering, University of Texas at Dallas, Richardson, TX 75080, USAAlthough there are well established object detection methods based on static images, their application to video data on a frame by frame basis faces two shortcomings: (i) lack of computational efficiency due to redundancy across image frames or by not using a temporal and spatial correlation of features across image frames, and (ii) lack of robustness to real-world conditions such as motion blur and occlusion. Since the introduction of the ImageNet Large Scale Visual Recognition Challenge (ILSVRC) in 2015, a growing number of methods have appeared in the literature on video object detection, many of which have utilized deep learning models. The aim of this paper is to provide a review of these papers on video object detection. An overview of the existing datasets for video object detection together with commonly used evaluation metrics is first presented. Video object detection methods are then categorized and a description of each of them is stated. Two comparison tables are provided to see their differences in terms of both accuracy and computational efficiency. Finally, some future trends in video object detection to address the challenges involved are noted.https://www.mdpi.com/2076-3417/10/21/7834video object detectionreview of video object detectiondeep learning-based video object detection |
spellingShingle | Haidi Zhu Haoran Wei Baoqing Li Xiaobing Yuan Nasser Kehtarnavaz A Review of Video Object Detection: Datasets, Metrics and Methods Applied Sciences video object detection review of video object detection deep learning-based video object detection |
title | A Review of Video Object Detection: Datasets, Metrics and Methods |
title_full | A Review of Video Object Detection: Datasets, Metrics and Methods |
title_fullStr | A Review of Video Object Detection: Datasets, Metrics and Methods |
title_full_unstemmed | A Review of Video Object Detection: Datasets, Metrics and Methods |
title_short | A Review of Video Object Detection: Datasets, Metrics and Methods |
title_sort | review of video object detection datasets metrics and methods |
topic | video object detection review of video object detection deep learning-based video object detection |
url | https://www.mdpi.com/2076-3417/10/21/7834 |
work_keys_str_mv | AT haidizhu areviewofvideoobjectdetectiondatasetsmetricsandmethods AT haoranwei areviewofvideoobjectdetectiondatasetsmetricsandmethods AT baoqingli areviewofvideoobjectdetectiondatasetsmetricsandmethods AT xiaobingyuan areviewofvideoobjectdetectiondatasetsmetricsandmethods AT nasserkehtarnavaz areviewofvideoobjectdetectiondatasetsmetricsandmethods AT haidizhu reviewofvideoobjectdetectiondatasetsmetricsandmethods AT haoranwei reviewofvideoobjectdetectiondatasetsmetricsandmethods AT baoqingli reviewofvideoobjectdetectiondatasetsmetricsandmethods AT xiaobingyuan reviewofvideoobjectdetectiondatasetsmetricsandmethods AT nasserkehtarnavaz reviewofvideoobjectdetectiondatasetsmetricsandmethods |