1. Convolutional neural networks Memory optimization Inference with Splitting Image
- Author
-
Tristan Hascoet, Yasuo Ariki, Tetsuya Takiguchi, Ryoichi Takashima, and Weihao Zhuang
- Subjects
Computer science ,business.industry ,Computer Science::Neural and Evolutionary Computation ,ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION ,Phase (waves) ,Inference ,Pattern recognition ,Convolutional neural network ,Image (mathematics) ,Memory management ,Computer Science::Computer Vision and Pattern Recognition ,Artificial intelligence ,Layer (object-oriented design) ,business - Abstract
In this paper, we propose a method to reduce the memory requirement of Convolutional Neural Networks (CNN) in the inference phase. Before feeding an input image into the CNN model, input image will split evenly to several sub-images and feed them into models respectively and combine the output after a certain layer.
- Published
- 2020
- Full Text
- View/download PDF