Convolution kernel adaptation to calibrated fisheye
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913221363892224 |
|---|---|
| author | Berenguel-Baeta, Bruno Santos-Villafranca, Maria Bermudez-Cameo, Jesus Perez-Yus, Alejandro Guerrero, Jose J. |
| author_facet | Berenguel-Baeta, Bruno Santos-Villafranca, Maria Bermudez-Cameo, Jesus Perez-Yus, Alejandro Guerrero, Jose J. |
| contents | Convolution kernels are the basic structural component of convolutional neural networks (CNNs). In the last years there has been a growing interest in fisheye cameras for many applications. However, the radially symmetric projection model of these cameras produces high distortions that affect the performance of CNNs, especially when the field of view is very large. In this work, we tackle this problem by proposing a method that leverages the calibration of cameras to deform the convolution kernel accordingly and adapt to the distortion. That way, the receptive field of the convolution is similar to standard convolutions in perspective images, allowing us to take advantage of pre-trained networks in large perspective datasets. We show how, with just a brief fine-tuning stage in a small dataset, we improve the performance of the network for the calibrated fisheye with respect to standard convolutions in depth estimation and semantic segmentation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_01456 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Convolution kernel adaptation to calibrated fisheye Berenguel-Baeta, Bruno Santos-Villafranca, Maria Bermudez-Cameo, Jesus Perez-Yus, Alejandro Guerrero, Jose J. Computer Vision and Pattern Recognition Convolution kernels are the basic structural component of convolutional neural networks (CNNs). In the last years there has been a growing interest in fisheye cameras for many applications. However, the radially symmetric projection model of these cameras produces high distortions that affect the performance of CNNs, especially when the field of view is very large. In this work, we tackle this problem by proposing a method that leverages the calibration of cameras to deform the convolution kernel accordingly and adapt to the distortion. That way, the receptive field of the convolution is similar to standard convolutions in perspective images, allowing us to take advantage of pre-trained networks in large perspective datasets. We show how, with just a brief fine-tuning stage in a small dataset, we improve the performance of the network for the calibrated fisheye with respect to standard convolutions in depth estimation and semantic segmentation. |
| title | Convolution kernel adaptation to calibrated fisheye |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2402.01456 |