Speeding up Convolutional Neural Networks with Low Rank Expansions
This addresses the deployability issue of CNNs for computer vision practitioners by providing a tunable speedup method, though it is incremental as it builds on existing low-rank approximation techniques.
The paper tackles the problem of speeding up convolutional neural networks, which are computationally demanding, by exploiting filter redundancy to construct low-rank bases, achieving up to 4.5x speedup with minimal accuracy loss in scene text character recognition.
The focus of this paper is speeding up the evaluation of convolutional neural networks. While delivering impressive results across a range of computer vision and machine learning tasks, these networks are computationally demanding, limiting their deployability. Convolutional layers generally consume the bulk of the processing time, and so in this work we present two simple schemes for drastically speeding up these layers. This is achieved by exploiting cross-channel or filter redundancy to construct a low rank basis of filters that are rank-1 in the spatial domain. Our methods are architecture agnostic, and can be easily applied to existing CPU and GPU convolutional frameworks for tuneable speedup performance. We demonstrate this with a real world network designed for scene text character recognition, showing a possible 2.5x speedup with no loss in accuracy, and 4.5x speedup with less than 1% drop in accuracy, still achieving state-of-the-art on standard benchmarks.