VisDrone-DET2018: The Vision Meets Drone Object Detection in Image Challenge Results

Pengfei Zhu, Longyin Wen, Dawei Du, Xiao Bian, Haibin Ling, Qinghua Hu, Qinqin Nie, Hao Cheng, Chenfeng Liu, Xiaoyu Liu, Wenya Ma, Haotian Wu, Lianjie Wang, Arne Schumann, Chase Brown, Chen Qian, Chengzheng Li, Dongdong Li, Emmanouil Michail, Fan Zhang, Feng Ni, Feng Zhu, Guanghui Wang, Haipeng Zhang, Han Deng, Hao Liu, Haoran Wang, Heqian Qiu, Honggang Qi, Honghui Shi, Hongliang Li, Hongyu Xu, Hu Lin, Ioannis Kompatsiaris, Jian Cheng, Jianqiang Wang, Jianxiu Yang, Jingkai Zhou, Juanping Zhao, K. J. Joseph, Kaiwen Duan, Karthik Suresh, Ke Bo, Ke Wang, Konstantinos Avgerinakis, Lars Sommer, Lei Zhang, Li Yang, Lin Cheng, Lin Ma, Liyu Lu, Lu Ding, Minyu Huang, Naveen Kumar Vedurupaka, Nehal Mamgain, Nitin Bansal, Oliver Acatay, Panagiotis Giannakeris, Qian Wang, Qijie Zhao, Qingming Huang, Qiong Liu, Qishang Cheng, Qiuchen Sun, Robert Laganire, Sheng Jiang, Shengjin Wang, Shubo Wei, Siwei Wang, Stefanos Vrochidis, Sujuan Wang, Tiaojio Lee, Usman Sajid, Vineeth N. Balasubramanian, Wei Li, Wei Zhang, Weikun Wu, Wenchi Ma, Wenrui He, Wenzhe Yang, Xiaoyu Chen, Xin Sun, Xinbin Luo, Xintao Lian, Xiufang Li, Yangliu Kuai, Yali Li, Yi Luo, Yifan Zhang, Yiling Liu, Ying Li, Yong Wang, Yongtao Wang, Yuanwei Wu, Yue Fan, Yunchao Wei, Yuqin Zhang, Zexin Wang, Zhangyang Wang, Zhaoyue Xia, Zhen Cui, Zhenwei He, Zhipeng Deng, Zhiyao Guo, Zichen Song; Proceedings of the European Conference on Computer Vision (ECCV) Workshops, 2018, pp. 0-0

Abstract


Object detection is a hot topic with various applications in computer vision, e.g., image understanding, autonomous driving, and video surveillance. Much of the progresses have been driven by the availability of object detection benchmark datasets, including PASCAL VOC, ImageNet, and MS COCO. However, object detection on the drone platform is still a challenging task, due to various factors such as view point change, occlusion, and scales. To narrow the gap between current object detection performance and the real-world requirements, we organized the Vision Meets Drone (VisDrone2018) Object Detection in Image challenge in conjunction with the 15th European Conference on Computer Vision (ECCV 2018). Specifically, we release a large-scale drone-based dataset, including 8,599 images (6,471 for training, 548 for validation, and 1,580 for testing) with rich annotations, including object bounding boxes, object categories, occlusion, truncation ratios, etc. Featuring a diverse real-world scenarios, the dataset was collected using various drone models, in different scenarios (across 14 different cities spanned over thousands of kilometres), and under various weather and lighting conditions. We mainly focus on ten object categories in object detection, i.e., pedestrian, person, car, van, bus, truck, motor, bicycle, awning-tricycle, and tricycle. Some rarely occurring special vehicles (e.g., machineshop truck, forklift truck, and tanker) are ignored in evaluation. The dataset is extremely challenging due to various factors, including large scale and pose variations, occlusion, and clutter background. We present the evaluation protocol of the VisDrone-DET2018 challenge and the comparison results of 38 detectors on the released dataset, which are publicly available on the challenge website: http://www.aiskyeye.com/. We expect the challenge to largely boost the research and development in object detection in images on drone platforms.

Related Material


[pdf]
[bibtex]
@InProceedings{Zhu_2018_ECCV_Workshops,
author = {Zhu, Pengfei and Wen, Longyin and Du, Dawei and Bian, Xiao and Ling, Haibin and Hu, Qinghua and Nie, Qinqin and Cheng, Hao and Liu, Chenfeng and Liu, Xiaoyu and Ma, Wenya and Wu, Haotian and Wang, Lianjie and Schumann, Arne and Brown, Chase and Qian, Chen and Li, Chengzheng and Li, Dongdong and Michail, Emmanouil and Zhang, Fan and Ni, Feng and Zhu, Feng and Wang, Guanghui and Zhang, Haipeng and Deng, Han and Liu, Hao and Wang, Haoran and Qiu, Heqian and Qi, Honggang and Shi, Honghui and Li, Hongliang and Xu, Hongyu and Lin, Hu and Kompatsiaris, Ioannis and Cheng, Jian and Wang, Jianqiang and Yang, Jianxiu and Zhou, Jingkai and Zhao, Juanping and Joseph, K. J. and Duan, Kaiwen and Suresh, Karthik and Bo, Ke and Wang, Ke and Avgerinakis, Konstantinos and Sommer, Lars and Zhang, Lei and Yang, Li and Cheng, Lin and Ma, Lin and Lu, Liyu and Ding, Lu and Huang, Minyu and Kumar Vedurupaka, Naveen and Mamgain, Nehal and Bansal, Nitin and Acatay, Oliver and Giannakeris, Panagiotis and Wang, Qian and Zhao, Qijie and Huang, Qingming and Liu, Qiong and Cheng, Qishang and Sun, Qiuchen and Laganire, Robert and Jiang, Sheng and Wang, Shengjin and Wei, Shubo and Wang, Siwei and Vrochidis, Stefanos and Wang, Sujuan and Lee, Tiaojio and Sajid, Usman and Balasubramanian, Vineeth N. and Li, Wei and Zhang, Wei and Wu, Weikun and Ma, Wenchi and He, Wenrui and Yang, Wenzhe and Chen, Xiaoyu and Sun, Xin and Luo, Xinbin and Lian, Xintao and Li, Xiufang and Kuai, Yangliu and Li, Yali and Luo, Yi and Zhang, Yifan and Liu, Yiling and Li, Ying and Wang, Yong and Wang, Yongtao and Wu, Yuanwei and Fan, Yue and Wei, Yunchao and Zhang, Yuqin and Wang, Zexin and Wang, Zhangyang and Xia, Zhaoyue and Cui, Zhen and He, Zhenwei and Deng, Zhipeng and Guo, Zhiyao and Song, Zichen},
title = {VisDrone-DET2018: The Vision Meets Drone Object Detection in Image Challenge Results},
booktitle = {Proceedings of the European Conference on Computer Vision (ECCV) Workshops},
month = {September},
year = {2018}
}