LAUSR

Perspective-Adaptive Convolutions for Scene Parsing

Sign Up to like & get
recommendations!
0 Published in 2020 at "IEEE Transactions on Pattern Analysis and Machine Intelligence"

DOI: 10.1109/tpami.2018.2890637

Abstract: Many existing scene parsing methods adopt Convolutional Neural Networks with receptive fields of fixed sizes and shapes, which frequently results in inconsistent predictions of large objects and invisibility of small objects. To tackle this issue,… read more here.

Keywords: convolutions scene; perspective adaptive; receptive fields; adaptive convolutions ... See more keywords