一种图像超分辨率神经网络参数效率提升的方法
By establishing a mapping relationship between convolutional layers and graph structures in an image super-resolution neural network, and using a random graph generator to guide sparsification and increase the number of channels, the problems of low parameter utilization efficiency and inconvenient deployment in existing technologies are solved, achieving efficient network parameter optimization and strong generalization.
CN116309060BActive Publication Date: 2026-07-17TSINGHUA SHENZHEN INTERNATIONAL GRADUATE SCHOOL
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- TSINGHUA SHENZHEN INTERNATIONAL GRADUATE SCHOOL
- Filing Date
- 2023-03-16
- Publication Date
- 2026-07-17
Smart Images
Figure CN116309060B_ABST
Abstract
一种图像超分辨率神经网络参数效率提升的方法，包括如下步骤：S1、建立神经网络的卷积层和图结构的映射关系；S2、通过随机图生成器生成不同的图结构；S3、在随机图中选用合适的结构点以指导网络卷积层结构的稀疏化过程；S4、将选取的图结构映射回神经网络的卷积层之中；S5、使用同样的图结构进行映射，应用至图像超分辨率神经网络的所有中间层中；网络的每一层都保持同样的子结构；S6、将图像超分辨率网络的通道数进行扩大；S7、进行模型的训练。本方法可以节约参数储存的空间，而且可以实现简单部署，计算开销少，可定义范围灵活，泛化性强，可以轻松实现大规模的应用。
Need to check novelty before this filing date? Find Prior Art