SpherePHD: Applying CNNs on a Spherical PolyHeDron Representation of 360° Images
| Autor: | Jongseob Yun, Wonjune Cho, Kuk-Jin Yoon, Jaeseok Jeong, Yeonkun Lee |
|---|---|
| Rok vydání: | 2019 |
| Předmět: |
Euclidean space
business.industry Computer science Deep learning 020206 networking & telecommunications 02 engineering and technology Convolutional neural network Image (mathematics) Convolution Distortion 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing Segmentation Angular resolution Computer vision Artificial intelligence Representation (mathematics) business Spherical polyhedron |
| Zdroj: | CVPR |
| DOI: | 10.1109/cvpr.2019.00940 |
| Popis: | Omni-directional cameras have many advantages overconventional cameras in that they have a much wider field-of-view (FOV). Accordingly, several approaches have beenproposed recently to apply convolutional neural networks(CNNs) to omni-directional images for various visual tasks.However, most of them use image representations defined inthe Euclidean space after transforming the omni-directionalviews originally formed in the non-Euclidean space. Thistransformation leads to shape distortion due to nonuniformspatial resolving power and the loss of continuity. Theseeffects make existing convolution kernels experience diffi-culties in extracting meaningful information. This paper presents a novel method to resolve such prob-lems of applying CNNs to omni-directional images. Theproposed method utilizes a spherical polyhedron to rep-resent omni-directional views. This method minimizes thevariance of the spatial resolving power on the sphere sur-face, and includes new convolution and pooling methodsfor the proposed representation. The proposed method canalso be adopted by any existing CNN-based methods. Thefeasibility of the proposed method is demonstrated throughclassification, detection, and semantic segmentation taskswith synthetic and real datasets. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|