Vision Meets Drones: A Challenge
arXiv:1804.07437
Abstract
In this paper we present a large-scale visual object detection and tracking benchmark, named VisDrone2018, aiming at advancing visual understanding tasks on the drone platform. The images and video sequences in the benchmark were captured over various urban/suburban areas of 14 different cities across China from north to south. Specifically, VisDrone2018 consists of 263 video clips and 10,209 images (no overlap with video clips) with rich annotations, including object bounding boxes, object categories, occlusion, truncation ratios, etc. With intensive amount of effort, our benchmark has more than 2.5 million annotated instances in 179,264 images/video frames. Being the largest such dataset ever published, the benchmark enables extensive evaluation and investigation of visual analysis algorithms on the drone platform. In particular, we design four popular tasks with the benchmark, including object detection in images, object detection in videos, single object tracking, and multi-object tracking. All these tasks are extremely challenging in the proposed dataset due to factors such as occlusion, large scale and pose variation, and fast motion. We hope the benchmark largely boost the research and development in visual analysis on drone platforms.
arXiv comment11 pages, 11 figures
Cited by in corpus (38)
- A Survey of Deep Learning-based Object Detection
- Object Detection in Aerial Images: A Large-Scale Benchmark and Challenges
- Deep Learning for UAV-based Object Detection and Tracking: A Survey
- SlimYOLOv3: Narrower, Faster and Better for Real-Time UAV Applications
- A Transformer-Based Feature Segmentation and Region Alignment Method For UAV-View Geo-Localization
- Detecting and Tracking Small and Dense Moving Objects in Satellite Videos: A Benchmark
- CNN-based Density Estimation and Crowd Counting: A Survey
- TPH-YOLOv5: Improved YOLOv5 Based on Transformer Prediction Head for Object Detection on Drone-captured Scenarios
- Focus-and-Detect: A Small Object Detection Framework for Aerial Images
- Semantics for Robotic Mapping, Perception and Interaction: A Survey
- Bridging the Gap Between Computational Photography and Visual Recognition
- E-UAV: An Edge-based Energy-Efficient Object Detection System for Unmanned Aerial Vehicles
- PENet: Object Detection using Points Estimation in Aerial Images
- BoLTVOS: Box-Level Tracking for Video Object Segmentation
- AdaZoom: Adaptive Zoom Network for Multi-Scale Object Detection in Large Scenes
- Clustered Object Detection in Aerial Images
- Vehicle Re-identification in Aerial Imagery: Dataset and Approach
- Efficient Object Detection Model for Real-Time UAV Applications
- Dense Crowds Detection and Counting with a Lightweight Architecture
- A Survey on Deep Domain Adaptation and Tiny Object Detection Challenges, Techniques and Datasets
- UAVid: A Semantic Segmentation Dataset for UAV Imagery
- Collision Detection: An Improved Deep Learning Approach Using SENet and ResNext
- University-1652: A Multi-view Multi-source Benchmark for Drone-based Geo-localization
- On Creating Benchmark Dataset for Aerial Image Interpretation: Reviews, Guidances and Million-AID
- Report on UG^2+ Challenge Track 1: Assessing Algorithms to Improve Video Object Detection and Classification from Unconstrained Mobility Platforms
- Straight to Shapes++: Real-time Instance Segmentation Made More Accurate
- Density Map Guided Object Detection in Aerial Images
- Crowd Counting on Images with Scale Variation and Isolated Clusters
- Multiple Object Tracking with Motion and Appearance Cues
- Kite: Automatic speech recognition for unmanned aerial vehicles
- Reliable Label Bootstrapping for Semi-Supervised Learning
- Multi-Drone based Single Object Tracking with Agent Sharing Network
- Congestion Analysis of Convolutional Neural Network-Based Pedestrian Counting Methods on Helicopter Footage
- Detection of Collision-Prone Vehicle Behavior at Intersections using Siamese Interaction LSTM
- Localizing Grouped Instances for Efficient Detection in Low-Resource Scenarios
- Robust Real-Time Pedestrian Detection on Embedded Devices
- Dense Multiscale Feature Fusion Pyramid Networks for Object Detection in UAV-Captured Images
- PANDA: A Gigapixel-level Human-centric Video Dataset