Hierarchical Convolutional Network
A convolutional network architecture that exploits the hierarchical vector-space structure of the image Gaussian jet and achieves scale invariance through a retraction operation, without explicit estimation of scale parameters, remaining robust to unknown Gaussian blur.
Timeline (last 90 days)
The paper proposes Graded Convolutional Networks, which achieve scale invariance without explicit scale parameters through a retraction operation.
Graded convolutional networks are expected to be applied to complex visual tasks where scale variation and blur are common, such as object recognition and medical image analysis.
All Facts (2)
The paper proposes Graded Convolutional Networks, which achieve scale invariance without explicit scale parameters through a retraction operation.
50%UnverifiedGraded convolutional networks are expected to be applied to complex visual tasks where scale variation and blur are common, such as object recognition and medical image analysis.
50%