Topics include clustering, sparse coding, autoencoders, restricted Boltzmann machines, and deep belief networks.
主题主要有聚类，稀疏编码，局限玻尔兹曼机和深度信念网络。
We expand it into two-layer structure based on sparse coding theory, and use Gaus-sian pyramid to solve scale problems.
我们修改基于特征的增量编码模型，根据稀疏编码原理，将其扩展成层次结构，并利用高斯金字塔解决尺度问题。
Image sparse representation data got by sparse decomposition method are compressed by quantization and coding.
在图像稀疏分解的基础上，对分解出的数据进行压缩。
Image sparse representation data got by sparse decomposition method are compressed by quantization and coding.
在图像稀疏分解的基础上，对分解出的数据进行压缩。
应用推荐