Fast Image Scanning with Deep Max-Pooling Convolutional Neural Networks
arXiv:1302.1700v1358 citations
Originality Highly original
AI Analysis
This addresses a critical bottleneck for real-time image analysis applications, offering a significant performance improvement.
The paper tackled the high computational cost of sliding window image scanning with deep neural networks by using dynamic programming, achieving speedups of orders of magnitude even with max-pooling layers.
Deep Neural Networks now excel at image classification, detection and segmentation. When used to scan images by means of a sliding window, however, their high computational complexity can bring even the most powerful hardware to its knees. We show how dynamic programming can speedup the process by orders of magnitude, even when max-pooling layers are present.