Next Article in Journal
Using Climate-Sensitive 3D City Modeling to Analyze Outdoor Thermal Comfort in Urban Areas
Next Article in Special Issue
Machine Learning-Based Processing Proof-of-Concept Pipeline for Semi-Automatic Sentinel-2 Imagery Download, Cloudiness Filtering, Classifications, and Updates of Open Land Use/Land Cover Datasets
Previous Article in Journal
Density-Based Spatial Clustering and Ordering Points Approach for Characterizations of Tourist Behaviour
Previous Article in Special Issue
Rainfall-Induced Shallow Landslide Susceptibility Mapping at Two Adjacent Catchments Using Advanced Machine Learning Algorithms
Article

Multi-View Instance Matching with Learned Geometric Soft-Constraints

1
EcoVision Lab, Photogrammetry and Remote Sensing Group, Institute of Geodesy and Photogrammetry, ETH Zurich, 8093 Zurich, Switzerland
2
IRISA, Université Bretagne Sud, 56000 Vannes, France
*
Author to whom correspondence should be addressed.
ISPRS Int. J. Geo-Inf. 2020, 9(11), 687; https://0-doi-org.brum.beds.ac.uk/10.3390/ijgi9110687
Received: 14 October 2020 / Revised: 3 November 2020 / Accepted: 11 November 2020 / Published: 18 November 2020
We present a new approach for matching urban object instances across multiple ground-level images for the ultimate goal of city-scale mapping of objects with high positioning accuracy. What makes this task challenging is the strong change in view-point, different lighting conditions, high similarity of neighboring objects, and variability in scale. We propose to turn object instance matching into a learning task, where image-appearance and geometric relationships between views fruitfully interact. Our approach constructs a Siamese convolutional neural network that learns to match two views of the same object given many candidate image cut-outs. In addition to image features, we propose utilizing location information about the camera and the object to support image evidence via soft geometric constraints. Our method is compared to existing patch matching methods to prove its edge over state-of-the-art. This takes us one step closer to the ultimate goal of city-wide object mapping from street-level imagery to benefit city administration. View Full-Text
Keywords: deep learning; siamese convolutional neural networks; urban object mapping deep learning; siamese convolutional neural networks; urban object mapping
Show Figures

Figure 1

MDPI and ACS Style

Nassar, A.S.; Lefèvre, S.; Wegner, J.D. Multi-View Instance Matching with Learned Geometric Soft-Constraints. ISPRS Int. J. Geo-Inf. 2020, 9, 687. https://0-doi-org.brum.beds.ac.uk/10.3390/ijgi9110687

AMA Style

Nassar AS, Lefèvre S, Wegner JD. Multi-View Instance Matching with Learned Geometric Soft-Constraints. ISPRS International Journal of Geo-Information. 2020; 9(11):687. https://0-doi-org.brum.beds.ac.uk/10.3390/ijgi9110687

Chicago/Turabian Style

Nassar, Ahmed S., Sébastien Lefèvre, and Jan D. Wegner 2020. "Multi-View Instance Matching with Learned Geometric Soft-Constraints" ISPRS International Journal of Geo-Information 9, no. 11: 687. https://0-doi-org.brum.beds.ac.uk/10.3390/ijgi9110687

Find Other Styles
Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.

Article Access Map by Country/Region

1
Back to TopTop