SaltiNet: Scan-path Prediction on 360 Degree Images using Saliency Volumes

July 11, 2017 · Entered Twilight · 🏛 2017 IEEE International Conference on Computer Vision Workshops (ICCVW)

"Last commit was 8.0 years ago (≥5 year threshold)"

Evidence collected by the PWNC Scanner

Repo contents: .gitignore, LICENSE, README.md, _config.yml, authors, figs, requirements.txt, src, utils.py

Authors Marc Assens, Kevin McGuinness, Xavier Giro-i-Nieto, Noel E. O'Connor arXiv ID 1707.03123 Category cs.CV: Computer Vision Cross-listed cs.MM Citations 129 Venue 2017 IEEE International Conference on Computer Vision Workshops (ICCVW) Repository https://github.com/massens/saliency-360salient-2017 ⭐ 58 Last Checked 1 month ago

Abstract

We introduce SaltiNet, a deep neural network for scanpath prediction trained on 360-degree images. The model 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 computed from a binary cross entropy (BCE) loss over downsampled versions of the saliency volumes. Sampling strategies over these volumes are used to generate scanpaths over the 360-degree images. Our experiments show the advantages of using saliency volumes, and how they can be used for related tasks. Our source code and trained models available at https://github.com/massens/saliency-360salient-2017.