Auto-DeepLab: Hierarchical Neural Architecture Search for Semantic Image Segmentation
Abstract
Recently, Neural Architecture Search (NAS) has successfully identified neural network architectures that exceed human designed ones on large-scale image classification. In this paper, we study NAS for semantic image segmentation. Existing works often focus on searching the repeatable cell structure, while hand-designing the outer network structure that controls the spatial resolution changes. This choice simplifies the search space, but becomes increasingly problematic for dense image prediction which exhibits a lot more network level architectural variations. Therefore, we propose to search the network level structure in addition to the cell level structure, which forms a hierarchical architecture search space. We present a network level search space that includes many popular designs, and develop a formulation that allows efficient gradient-based architecture search (3 P100 GPU days on Cityscapes images). We demonstrate the effectiveness of the proposed method on the challenging Cityscapes, PASCAL VOC 2012, and ADE20K datasets. Auto-DeepLab, our architecture searched specifically for semantic image segmentation, attains state-of-the-art performance without any ImageNet pretraining.
Cite
@article{arxiv.1901.02985,
title = {Auto-DeepLab: Hierarchical Neural Architecture Search for Semantic Image Segmentation},
author = {Chenxi Liu and Liang-Chieh Chen and Florian Schroff and Hartwig Adam and Wei Hua and Alan Yuille and Li Fei-Fei},
journal= {arXiv preprint arXiv:1901.02985},
year = {2019}
}
Comments
To appear in CVPR 2019 as oral. Code for Auto-DeepLab released at https://github.com/tensorflow/models/tree/master/research/deeplab