UPRIGHT ADJUSTMENT WITH GRAPH CONVOLUTIONAL NETWORKS

Cited 0 time in webofscience Cited 0 time in scopus
  • Hit : 20
  • Download : 0
We present a novel method for the upright adjustment of 360 degrees. images. Our network consists of two modules, which are a convolutional neural network (CNN) and a graph convolutional network (GCN). The input 360 degrees. images is processed with the CNN for visual feature extraction, and the extracted feature map is converted into a graph that finds a spherical representation of the input. We also introduce a novel loss function to address the issue of discrete probability distributions defined on the surface of a sphere. Experimental results demonstrate that our method outperforms fully connected-based methods.
Publisher
IEEE
Issue Date
2020-09
Language
English
Citation

2020 IEEE International Conference on Image Processing, ICIP 2020, pp.1058 - 1062

ISSN
1522-4880
DOI
10.1109/ICIP40778.2020.9190715
URI
http://hdl.handle.net/10203/288300
Appears in Collection
RIMS Conference Papers
Files in This Item
There are no files associated with this item.

qr_code

  • mendeley

    citeulike


rss_1.0 rss_2.0 atom_1.0