Skip to content

This is the official repository for “Drone-based RGB-Infrared Cross-Modality Vehicle Detection via Uncertainty-Aware Learning” (IEEE T-CSVT 2022).

License

Notifications You must be signed in to change notification settings

SunYM2020/UA-CMDet

Repository files navigation

UA-CMDet

This is the official repository for “Drone-based RGB-Infrared Cross-Modality Vehicle Detection via Uncertainty-Aware Learning” (IEEE T-CSVT 2022). The repo is based on AerialDetection and mmdetection.

Abstract

Drone-based vehicle detection aims at detecting vehicle locations and categories in aerial images. It empowers smart city traffic management and disaster relief. Researchers have made a great deal of effort in this area and achieved considerable progress. However, because of the paucity of data under extreme conditions, drone-based vehicle detection remains a challenge when objects are difficult to distinguish, particularly in low-light conditions. To fill this gap, we constructed a large-scale drone-based RGB-infrared vehicle detection dataset called DroneVehicle, which contains 28, 439 RGB-infrared image pairs covering urban roads, residential areas, parking lots, and other scenarios from day to night. Cross-modal images provide complementary information for vehicle detection, but also introduce redundant information. To handle this dilemma, we further propose an uncertainty-aware cross-modality vehicle detection (UA-CMDet) framework to improve detection performance in complex environments. Specifically, we design an uncertainty-aware module using cross-modal intersection over union and illumination estimation to quantify the uncertainty of each object. Our method takes uncertainty as a weight to boost model learning more effectively while reducing bias caused by high-uncertainty objects. For more robust cross-modal integration, we further perform illumination-aware non-maximum suppression during inference. Extensive experiments on our DroneVehicle and two challenging RGB-infrared object detection datasets demonstrated the advanced flexibility and superior performance of UA-CMDet over competing methods.

framework

Requirements

  • Ubuntu 18.04
  • CUDA 10.0
  • NCCL 2+
  • GCC 4.9+
  • PyTorch 1.1
  • Python 3.7
  • torchvision 0.3
  • mmcv 0.4.3

Installation

Please refer to INSTALL.md for installation.

Getting Started

Train with a single GPU.

python tools/train.py configs/DroneVehicle/UACMDet.py

Inference

python tools/test.py configs/DroneVehicle/UACMDet.py work_dirs/UACMDet/${CHECKPOINT_FILE} --out work_dirs/UACMDet/results.pkl

Parse the results.pkl

python tools/parse_results.py --config configs/DroneVehicle/UACMDet.py --type OBB

Evaluation

python eval/DroneVehicleEval.py

DroneVehicle Dataset

  1. DroneVehicle dataset is available now.
  2. Annotation files in COCO format can be downloaded in the link. (extracting code: vdbt)

Citation

Please cite this paper if you want to use it in your work.

@ARTICLE{sun2020drone,
  title={Drone-based RGB-Infrared Cross-Modality Vehicle Detection via Uncertainty-Aware Learning}, 
  author={Sun, Yiming and Cao, Bing and Zhu, Pengfei and Hu, Qinghua},
  journal={IEEE Transactions on Circuits and Systems for Video Technology}, 
  year={2022},
  volume={32},
  number={10},
  pages={6700-6713},
  doi={10.1109/TCSVT.2022.3168279}
}

About

This is the official repository for “Drone-based RGB-Infrared Cross-Modality Vehicle Detection via Uncertainty-Aware Learning” (IEEE T-CSVT 2022).

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published