Back to Search
Start Over
Multi-View Instance Matching with Learned Geometric Soft-Constraints
- Source :
- ISPRS International Journal of Geo-Information, ISPRS International Journal of Geo-Information, MDPI, 2020, 9 (11), pp.687. ⟨10.3390/ijgi9110687⟩, Volume 9, Issue 11, ISPRS International Journal of Geo-Information, 9 (11), ISPRS International Journal of Geo-Information, 2020, 9 (11), pp.687. ⟨10.3390/ijgi9110687⟩, ISPRS International Journal of Geo-Information, Vol 9, Iss 687, p 687 (2020)
- Publication Year :
- 2020
- Publisher :
- HAL CCSD, 2020.
-
Abstract
- 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.<br />ISPRS International Journal of Geo-Information, 9 (11)<br />ISSN:2220-9964
- Subjects :
- Matching (statistics)
Similarity (geometry)
Computer science
Geography, Planning and Development
0211 other engineering and technologies
lcsh:G1-922
02 engineering and technology
Convolutional neural network
siamese convolutional neural networks
Task (project management)
Image (mathematics)
0202 electrical engineering, electronic engineering, information engineering
Earth and Planetary Sciences (miscellaneous)
deep learning
urban object mapping
Computer vision
Computers in Earth Sciences
021101 geological & geomatics engineering
business.industry
Deep learning
Object (computer science)
[INFO.INFO-TI]Computer Science [cs]/Image Processing [eess.IV]
020201 artificial intelligence & image processing
Artificial intelligence
business
Scale (map)
lcsh:Geography (General)
Subjects
Details
- Language :
- English
- ISSN :
- 22209964
- Database :
- OpenAIRE
- Journal :
- ISPRS International Journal of Geo-Information, ISPRS International Journal of Geo-Information, MDPI, 2020, 9 (11), pp.687. ⟨10.3390/ijgi9110687⟩, Volume 9, Issue 11, ISPRS International Journal of Geo-Information, 9 (11), ISPRS International Journal of Geo-Information, 2020, 9 (11), pp.687. ⟨10.3390/ijgi9110687⟩, ISPRS International Journal of Geo-Information, Vol 9, Iss 687, p 687 (2020)
- Accession number :
- edsair.doi.dedup.....0df4adfdd8a831cec2e7551a51f90f61