Bharath Hariharan, Pablo Arbeláez, Ross Girshick, Jitendra Malik
TL;DR本研究提出了一种名为 Simultaneous Detection and Segmentation 的任务，使用卷积神经网络实现目标检测、区域分割及语义分割，与现有方法相比具有更高的准确性，并提供了诊断工具以指导未来工作。
Abstract
We aim to detect all instances of a category in an image and, for each instance, mark the pixels that belong to it. We call this task simultaneous detection and segmentation (SDS). Unlike classical bounding box detection, SDS requires a segmentation and not just a box. Unlike classical