TY - JOUR

T1 - LR-CNN

T2 - 2020 24th ISPRS Congress on Technical Commission II

AU - Liao, Liao

AU - Chen, Xiang

AU - Yang, Jingfeng

AU - Roth, Stefan

AU - Goesele, Michael

AU - Yang, Michael Ying

AU - Rosenhahn, Bodo

N1 - Funding information: This work was supported by German Research Foundation (DFG) grants COVMAP (RO 2497/12-2) and PhoenixD (EXC 2122, Project ID 390833453).

PY - 2020/8/3

Y1 - 2020/8/3

N2 - 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.

AB - 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.

KW - Deep Learning

KW - Feature Enhancement

KW - Object Detection

KW - Twin Region Proposal

KW - Vehicle Detection

UR - http://www.scopus.com/inward/record.url?scp=85091068652&partnerID=8YFLogxK

U2 - 10.5194/isprs-annals-V-2-2020-381-2020

DO - 10.5194/isprs-annals-V-2-2020-381-2020

M3 - Conference article

AN - SCOPUS:85091068652

VL - 5

SP - 381

EP - 388

JO - ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences

JF - ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences

SN - 2194-9042

IS - 2

Y2 - 31 August 2020 through 2 September 2020

ER -