Deformable ConvNets v2: More Deformable, Better Results

CVPR, Volume abs/1811.11168, 2019, Pages 9308-9316.

Cited by: 166|Bibtex|Views104
EI
Other Links: dblp.uni-trier.de|academic.microsoft.com|arxiv.org

Abstract:

The superior performance of Deformable Convolutional Networks arises from its ability to adapt to the geometric variations of objects. Through an examination of its adaptive behavior, we observe that while the spatial support for its neural features conforms more closely than regular ConvNets to object structure, this support may neverthe...More

Code:

Data:

Full Text
Your rating :
0

 

Tags
Comments