Peter Corcoran, co-founder of FotoNation (now a core business unit of Xperi) and lead principle investigator and director of C3Imaging (a research partnership between Xperi and the National University of Ireland, Galway), presents the “Hybrid Semi-Parallel Deep Neural Networks (SPDNN) – Example Methodologies & Use Cases” tutorial at the May 2018 Embedded Vision Summit.
Deep neural networks (DNNs) are typically trained on specific datasets, optimized with particular discriminating capabilities. Often several different DNN topologies are developed solving closely related aspects of a computer vision problem. But to utilize these topologies together, leveraging their individual discriminating capabilities, requires implementing each DNN separately, increasing the cost of practical solutions.
In this talk, Corcoran develops a methodology to merge multiple deep networks using graph contraction. The resultant single network topology achieves a significant reduction in size over the individual networks. More significantly, this merged SPDNN network can be re-trained across the combined datasets used to train the original networks, improving its accuracy over the original networks. The result is a single network that is more generic, but with equivalent – or often enhanced – performance over a wider range of input data.
Examples of several problems in contemporary computer vision are solved using SPDNNs. These include significantly improving segmentation accuracy of eye-iris regions (a key component of iris biometric authentication) and mapping depth from monocular images, demonstrating equivalent performance to stereo depth mapping.