RoTIR: Rotation-Equivariant Network and Transformers for Fish Scale Image Registration
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866909270053748736 |
|---|---|
| author | Wang, Ruixiong Achim, Alin Raele-Rolfe, Renata Tong, Qiao Bergen, Dylan Hammond, Chrissy Cross, Stephen |
| author_facet | Wang, Ruixiong Achim, Alin Raele-Rolfe, Renata Tong, Qiao Bergen, Dylan Hammond, Chrissy Cross, Stephen |
| contents | Image registration is an essential process for aligning features of interest from multiple images. With the recent development of deep learning techniques, image registration approaches have advanced to a new level. In this work, we present 'Rotation-Equivariant network and Transformers for Image Registration' (RoTIR), a deep-learning-based method for the alignment of fish scale images captured by light microscopy. This approach overcomes the challenge of arbitrary rotation and translation detection, as well as the absence of ground truth data. We employ feature-matching approaches based on Transformers and general E(2)-equivariant steerable CNNs for model creation. Besides, an artificial training dataset is employed for semi-supervised learning. Results show RoTIR successfully achieves the goal of fish scale image registration. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2401_11270 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | RoTIR: Rotation-Equivariant Network and Transformers for Fish Scale Image Registration Wang, Ruixiong Achim, Alin Raele-Rolfe, Renata Tong, Qiao Bergen, Dylan Hammond, Chrissy Cross, Stephen Image and Video Processing Image registration is an essential process for aligning features of interest from multiple images. With the recent development of deep learning techniques, image registration approaches have advanced to a new level. In this work, we present 'Rotation-Equivariant network and Transformers for Image Registration' (RoTIR), a deep-learning-based method for the alignment of fish scale images captured by light microscopy. This approach overcomes the challenge of arbitrary rotation and translation detection, as well as the absence of ground truth data. We employ feature-matching approaches based on Transformers and general E(2)-equivariant steerable CNNs for model creation. Besides, an artificial training dataset is employed for semi-supervised learning. Results show RoTIR successfully achieves the goal of fish scale image registration. |
| title | RoTIR: Rotation-Equivariant Network and Transformers for Fish Scale Image Registration |
| topic | Image and Video Processing |
| url | https://arxiv.org/abs/2401.11270 |