PathGAN: visual scanpath prediction with generative adversarial networks

View/Open
Cita com:
hdl:2117/130229
Document typeConference lecture
Defense date2019
PublisherSpringer
Rights accessOpen Access
All rights reserved. This work is protected by the corresponding intellectual and industrial
property rights. Without prejudice to any existing legal exemptions, reproduction, distribution, public
communication or transformation of this work are prohibited without permission of the copyright holder
Abstract
We introduce PathGAN, a deep neural network for visual scanpath prediction trained on adversarial examples. A visual scanpath is defined as the sequence of fixation points over an image defined by a human observer with its gaze. PathGAN is composed of two parts, the generator and the discriminator. Both parts extract features from images using off-the-shelf networks, and train recurrent layers to generate or discriminate scanpaths accordingly. In scanpath prediction, the stochastic nature of the data makes it very difficult to generate realistic predictions using supervised learning strategies, but we adopt adversarial training as a suitable alternative. Our experiments prove how PathGAN improves the state of the art of visual scanpath prediction on the iSUN and Salient360! datasets.
Description
“This is a post-peer-review, pre-copyedit version of an article published in: Computer Vision – ECCV 2018 Workshops. The final authenticated version is available online at: http://dx.doi.org/10.1007/978-3-030-11021-5_25”.
CitationAssens, M. [et al.]. PathGAN: visual scanpath prediction with generative adversarial networks. A: Workshop on Egocentric Perception, Interaction and Computing. "Computer Vision: ECCV 2018 Workshops, Munich, Germany, September 8-14, 2018: proceedings, part V". Berlín: Springer, 2019, p. 406-422.
ISBN978-3-030-11021-5
Publisher versionhttps://link.springer.com/chapter/10.1007%2F978-3-030-11021-5_25
Other identifiershttps://imatge-upc.github.io/pathgan/
Files | Description | Size | Format | View |
---|---|---|---|---|
1809.00567.pdf | 3,781Mb | View/Open |