A Proper Orthogonal Decomposition approach for parameters reduction of Single Shot Detector networks
This work addresses real-time performance and training speed issues in object detection for embedded industrial systems, but it is incremental as it applies an existing reduction technique to a specific network.
The authors tackled the problem of reducing hyperparameters in Single Shot Detector networks to improve efficiency for resource-constrained systems, achieving a reduction in network dimension and a remarkable speedup in fine-tuning using the PASCAL VOC dataset.
As a major breakthrough in artificial intelligence and deep learning, Convolutional Neural Networks have achieved an impressive success in solving many problems in several fields including computer vision and image processing. Real-time performance, robustness of algorithms and fast training processes remain open problems in these contexts. In addition object recognition and detection are challenging tasks for resource-constrained embedded systems, commonly used in the industrial sector. To overcome these issues, we propose a dimensionality reduction framework based on Proper Orthogonal Decomposition, a classical model order reduction technique, in order to gain a reduction in the number of hyperparameters of the net. We have applied such framework to SSD300 architecture using PASCAL VOC dataset, demonstrating a reduction of the network dimension and a remarkable speedup in the fine-tuning of the network in a transfer learning context.