Skip to content

Deformable Convolutional Networks

Jaewon Chung edited this page Jan 7, 2019 · 4 revisions

Resources

Abstract

기존 CNN 구조의 한계:

Convolutional neural networks (CSSs) are inherently limited to model geometric transformations due to the fixed geometric structures in their building modules.

논문에서 제시하는 해결 방법:

In this work, we introduce two new modules ..., deformable convolution and deformable RoI pooling.

Introduction