Leveraging Synthetic Data in Object Detection on Unmanned Aerial Vehicles

22 Dec 2021  ·  Benjamin Kiefer, David Ott, Andreas Zell ·

Acquiring data to train deep learning-based object detectors on Unmanned Aerial Vehicles (UAVs) is expensive, time-consuming and may even be prohibited by law in specific environments. On the other hand, synthetic data is fast and cheap to access. In this work, we explore the potential use of synthetic data in object detection from UAVs across various application environments. For that, we extend the open-source framework DeepGTAV to work for UAV scenarios. We capture various large-scale high-resolution synthetic data sets in several domains to demonstrate their use in real-world object detection from UAVs by analyzing multiple training strategies across several models. Furthermore, we analyze several different data generation and sampling parameters to provide actionable engineering advice for further scientific research. The DeepGTAV framework is available at https://git.io/Jyf5j.

PDF Abstract

Datasets


Results from the Paper


 Ranked #1 on Object Detection on SeaDronesSee (using extra training data)

     Get a GitHub badge
Task Dataset Model Metric Name Metric Value Global Rank Uses Extra
Training Data
Benchmark
Object Detection SeaDronesSee Synth Pretrained Faster R-CNN ResNeXt-101-FPN mAP@0.5 59.20 # 1
Object Detection SeaDronesSee Yolo 5 mAP@0.5 54.74 # 3
Object Detection SeaDronesSee Synth Pretrained Yolo5 mAP@0.5 59.08 # 2
Object Detection SeaDronesSee Synth Pretrained EffDetD0 mAP@0.5 38.74 # 6

Methods


No methods listed for this paper. Add relevant methods here