LR-CNN: Local-aware region cnn for vehicle detection in aerial imagery

Wentong Liao, X. Chen, J. Yang, S. Roth, M. Goesele, M. Y. Yang, Bodo Rosenhahn

Research output: Chapter in Book/Report/Conference proceedingChapterAcademicpeer-review

8 Downloads (Pure)

Abstract

State-of-the-art object detection approaches such as Fast/Faster R-CNN, SSD, or YOLO have difficulties detecting dense, small targets with arbitrary orientation in large aerial images. The main reason is that using interpolation to align RoI features can result in a lack of accuracy or even loss of location information. We present the Local-aware Region Convolutional Neural Network (LR-CNN), a novel two-stage approach for vehicle detection in aerial imagery. We enhance translation invariance to detect dense vehicles and address the boundary quantization issue amongst dense vehicles by aggregating the high-precision RoIs’ features. Moreover, we resample high-level semantic pooled features, making them regain location information from the features of a shallower convolutional block. This strengthens the local feature invariance for the resampled features and enables detecting vehicles in an arbitrary orientation. The local feature invariance enhances the learning ability of the focal loss function, and the focal loss further helps to focus on the hard examples. Taken together, our method better addresses the challenges of aerial imagery. We evaluate our approach on several challenging datasets (VEDAI, DOTA), demonstrating a significant improvement over state-of-the-art methods. We demonstrate the good generalization ability of our approach on the DLR 3K dataset.
Original languageEnglish
Title of host publicationISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Subtitle of host publicationXXIV ISPRS Congress
EditorsN. Paparoditis, C. Mallet, F. Lafarge, F. Remondino, I. Toschi, T. Fuse
Place of PublicationNice
PublisherInternational Society for Photogrammetry and Remote Sensing (ISPRS)
Pages381-388
Number of pages8
VolumeV-2-2020
DOIs
Publication statusPublished - 3 Aug 2020
EventXXIVth ISPRS Congress 2020 - Nice-Acropolis Congress and Exhibition Centre, Nice, France
Duration: 4 Jul 202010 Jul 2020
Conference number: 24
http://www.isprs2020-nice.com

Publication series

NameISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
PublisherCopernicus
ISSN (Print)2194-9042

Conference

ConferenceXXIVth ISPRS Congress 2020
Abbreviated titleISPRS 2020
CountryFrance
CityNice
Period4/07/2010/07/20
Internet address

Fingerprint Dive into the research topics of 'LR-CNN: Local-aware region cnn for vehicle detection in aerial imagery'. Together they form a unique fingerprint.

  • Cite this

    Liao, W., Chen, X., Yang, J., Roth, S., Goesele, M., Yang, M. Y., & Rosenhahn, B. (2020). LR-CNN: Local-aware region cnn for vehicle detection in aerial imagery. In N. Paparoditis, C. Mallet, F. Lafarge, F. Remondino, I. Toschi, & T. Fuse (Eds.), ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences: XXIV ISPRS Congress (Vol. V-2-2020, pp. 381-388). (ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences). Nice: International Society for Photogrammetry and Remote Sensing (ISPRS). https://doi.org/10.5194/isprs-annals-V-2-2020-381-2020