AdaptIS: 自适应实例选择网络

Sep, 2019

AdaptIS: Adaptive Instance Selection Network

Konstantin Sofiiuk, Olga Barinova, Anton Konushin

TL;DR本文提出 Adaptive Instance Selection 网络结构，用于无类别实例分割，在与语义分割流水线相结合的情况下，可以轻松地实现全景分割，并在 Cityscapes 和 Mapillary 数据集上获得最先进的结果。

Abstract

We present adaptive instance selection network architecture for class-agnostic instance segmentation. Given an input image and a point $(x, y)$, it generates a mask for the object located at $(x, y)$. The network