Details
| Original language | English |
|---|---|
| Pages (from-to) | 58-74 |
| Number of pages | 17 |
| Journal | Geo-Spatial Information Science |
| Volume | 24 |
| Issue number | 1 |
| Early online date | 17 Nov 2020 |
| Publication status | Published - 2021 |
Abstract
In feature based image matching, distinctive features in images are detected and represented by feature descriptors. Matching is then carried out by assessing the similarity of the descriptors of potentially conjugate points. In this paper, we first shortly discuss the general framework. Then, we review feature detection as well as the determination of affine shape and orientation of local features, before analyzing feature description in more detail. In the feature description review, the general framework of local feature description is presented first. Then, the review discusses the evolution from hand-crafted feature descriptors, e.g. SIFT (Scale Invariant Feature Transform), to machine learning and deep learning based descriptors. The machine learning models, the training loss and the respective training data of learning-based algorithms are looked at in more detail; subsequently the various advantages and challenges of the different approaches are discussed. Finally, we present and assess some current research directions before concluding the paper.
Keywords
- affine shape estimation, descriptor learning, feature orientation, Image matching, image orientation
ASJC Scopus subject areas
- Social Sciences(all)
- Geography, Planning and Development
- Earth and Planetary Sciences(all)
- Computers in Earth Sciences
Cite this
- Standard
- Harvard
- Apa
- Vancouver
- BibTeX
- RIS
In: Geo-Spatial Information Science, Vol. 24, No. 1, 2021, p. 58-74.
Research output: Contribution to journal › Article › Research › peer review
}
TY - JOUR
T1 - Feature detection and description for image matching
T2 - from hand-crafted design to deep learning
AU - Chen, Lin
AU - Rottensteiner, Franz
AU - Heipke, Christian
N1 - Funding Information: The authors would like to thank NVIDIA Corp. for donating the GPU used in this research through its GPU grant program. The first author Lin Chen would also like to thank the China Scholarship Council (CSC) for financially supporting his PhD study.
PY - 2021
Y1 - 2021
N2 - In feature based image matching, distinctive features in images are detected and represented by feature descriptors. Matching is then carried out by assessing the similarity of the descriptors of potentially conjugate points. In this paper, we first shortly discuss the general framework. Then, we review feature detection as well as the determination of affine shape and orientation of local features, before analyzing feature description in more detail. In the feature description review, the general framework of local feature description is presented first. Then, the review discusses the evolution from hand-crafted feature descriptors, e.g. SIFT (Scale Invariant Feature Transform), to machine learning and deep learning based descriptors. The machine learning models, the training loss and the respective training data of learning-based algorithms are looked at in more detail; subsequently the various advantages and challenges of the different approaches are discussed. Finally, we present and assess some current research directions before concluding the paper.
AB - In feature based image matching, distinctive features in images are detected and represented by feature descriptors. Matching is then carried out by assessing the similarity of the descriptors of potentially conjugate points. In this paper, we first shortly discuss the general framework. Then, we review feature detection as well as the determination of affine shape and orientation of local features, before analyzing feature description in more detail. In the feature description review, the general framework of local feature description is presented first. Then, the review discusses the evolution from hand-crafted feature descriptors, e.g. SIFT (Scale Invariant Feature Transform), to machine learning and deep learning based descriptors. The machine learning models, the training loss and the respective training data of learning-based algorithms are looked at in more detail; subsequently the various advantages and challenges of the different approaches are discussed. Finally, we present and assess some current research directions before concluding the paper.
KW - affine shape estimation
KW - descriptor learning
KW - feature orientation
KW - Image matching
KW - image orientation
UR - http://www.scopus.com/inward/record.url?scp=85096179529&partnerID=8YFLogxK
U2 - 10.1080/10095020.2020.1843376
DO - 10.1080/10095020.2020.1843376
M3 - Article
AN - SCOPUS:85096179529
VL - 24
SP - 58
EP - 74
JO - Geo-Spatial Information Science
JF - Geo-Spatial Information Science
SN - 1009-5020
IS - 1
ER -