Neural algebra of classifiers
Date
2018-03-15
Authors
Santa Cruz, Rodrigo
Fernando, Basura
Cherian, Anoop
Gould, Stephen
Journal Title
Journal ISSN
Volume Title
Publisher
IEEE
Abstract
The world is fundamentally compositional, so it is natural to think of visual recognition as the recognition of basic visually primitives that are composed according to well-defined rules. This strategy allows us to recognize unseen complex concepts from simple visual primitives. However, the current trend in visual recognition follows a data greedy approach where huge amounts of data are required to learn models for any desired visual concept. In this paper, we build on the compositionality principle and develop an "algebra" to compose classifiers for complex visual concepts. To this end, we learn neural network modules to perform boolean algebra operations on simple visual classifiers. Since these modules form a complete functional set, a classifier for any complex visual concept defined as a boolean expression of primitives can be obtained by recursively applying the learned modules, even if we do not have a single training sample. As our experiments show, using such a framework, we can compose classifiers for complex visual concepts outperforming standard baselines on two well-known visual recognition benchmarks. Finally, we present a qualitative analysis of our method and its properties.
Description
Keywords
Citation
Collections
Source
Proceedings - 2018 IEEE Winter Conference on Applications of Computer Vision, WACV 2018
Type
Conference paper
Book Title
Entity type
Access Statement
License Rights
Restricted until
2037-12-31