Loading...
Loading...

Go to the content (press return)

Scanpath and saliency prediction on 360 degree images

Author
Assens, M.; Giro, X.; McGuinness, K.; O'Connor, N.
Type of activity
Journal article
Journal
Signal processing: image communication
Date of publication
2018-06-23
Volume
69
First page
8
Last page
14
DOI
https://doi.org/10.1016/j.image.2018.06.006 Open in new window
Project funding
Multimodal Signal Processing and Machine Learning on Graphs
Repository
http://hdl.handle.net/2117/119346 Open in new window
https://imatge.upc.edu/web/publications/scanpath-and-saliency-prediction-360-degree-images Open in new window
URL
https://www.sciencedirect.com/science/article/pii/S0923596518306209 Open in new window
Abstract
We introduce deep neural networks for scanpath and saliency prediction trained on 360-degree images. The scanpath prediction model called SaltiNet is based on a temporal-aware novel representation of saliency information named the saliency volume. The first part of the network consists of a model trained to generate saliency volumes, whose parameters are fit by back-propagation using a binary cross entropy (BCE) loss over downsampled versions of the saliency volumes. Sampling strategies over the...
Citation
Assens, M., Giro, X., McGuinness, K., O'Connor, N. Scanpath and saliency prediction on 360 degree images. "Signal processing: image communication", 23 Juny 2018, vol. 69, p. 8-14.
Keywords
deep learning, machine learning, saliency, scanpath, visual attention
Group of research
GPI - Image and Video Processing Group
IDEAI-UPC - Intelligent Data Science and Artificial Intelligence Research Center

Participants

  • Assens Reina, Marc  (author)
  • Giro Nieto, Xavier  (author)
  • McGuinness, Kevin  (author)
  • O'Connor, Noel  (author)