Abstract
Hierarchical feature learning based on convolutional neural networks (CNN) has recently shown significant potential in various computer vision tasks. While allowing high-quality discriminative feature learning, the downside of CNNs is the lack of explicit structure in features, which often leads to overfitting, absence of reconstruction from partial observations and limited generative abilities. Explicit structure is inherent in hierarchical compositional models, however, these lack the ability to optimize a well-defined cost function. We propose a novel analytic model of a basic unit in a layered hierarchical model with both explicit compositional structure and a well-defined discriminative
cost function. Our experiments on two datasets show that the proposed compositional model performs on a par with standard CNNs on discriminative tasks, while, due to explicit modeling of the structure in the feature units, affording a straight-forward visualization of parts and faster inference due to separability of
the units.
cost function. Our experiments on two datasets show that the proposed compositional model performs on a par with standard CNNs on discriminative tasks, while, due to explicit modeling of the structure in the feature units, affording a straight-forward visualization of parts and faster inference due to separability of
the units.
Original language | English |
---|---|
Title of host publication | Proceedings ICPR 2016 |
Subtitle of host publication | 23rd International Conference on Pattern Recognition |
Publisher | IEEE Computer Society |
Number of pages | 6 |
ISBN (Electronic) | 978-1-5090-4847-2 |
ISBN (Print) | 978-1-5090-4848-9 |
DOIs | |
Publication status | Published - 24 Apr 2017 |
Event | 23rd International Conference on Pattern Recognition - Cancun, Mexico Duration: 4 Dec 2016 → 8 Dec 2016 |
Conference
Conference | 23rd International Conference on Pattern Recognition |
---|---|
Country/Territory | Mexico |
City | Cancun |
Period | 4/12/16 → 8/12/16 |