ePrints@IIScePrints@IISc Home | About | Browse | Latest Additions | Advanced Search | Contact | Help

Machine vision benefits from human contextual expectations

Katti, Harish and Peelen, Marius and Arun, SP (2019) Machine vision benefits from human contextual expectations. In: SCIENTIFIC REPORTS, 9 .

[img]
Preview
PDF
Sci_ Rep_9_2019.pdf

Download (1MB) | Preview
Official URL: https://doi.org/10.1038/s41598-018-38427-0

Abstract

Scene context is known to facilitate object recognition in both machines and humans, suggesting that the underlying representations may be similar. Alternatively, they may be qualitatively different since the training experience of machines and humans are strikingly different. Machines are typically trained on images containing objects and their context, whereas humans frequently experience scenes without objects (such as highways without cars). If these context representations are indeed different, machine vision algorithms will be improved on augmenting them with human context representations, provided these expectations can be measured and are systematic. Here, we developed a paradigm to measure human contextual expectations. We asked human subjects to indicate the scale, location and likelihood at which cars or people might occur in scenes without these objects. This yielded highly systematic expectations that we could then accurately predict using scene features. This allowed us to predict human expectations on novel scenes without requiring explicit measurements. Next we augmented decisions made by deep neural networks with these predicted human expectations and obtained substantial gains in accuracy for detecting cars and people (1-3%) as well as on detecting associated objects (3-20%). In contrast, augmenting deep network decisions with other conventional computer vision features yielded far smaller gains. Taken together, our results show that augmenting deep neural networks with human-derived contextual expectations improves their performance, suggesting that contextual representations are qualitatively different in humans and deep neural networks.

Item Type: Journal Article
Additional Information: Copyright for this article belongs to Nature Publishing Group
Department/Centre: Division of Biological Sciences > Centre for Neuroscience
Depositing User: Id for Latest eprints
Date Deposited: 04 Mar 2019 06:00
Last Modified: 04 Mar 2019 06:00
URI: http://eprints.iisc.ac.in/id/eprint/61876

Actions (login required)

View Item View Item