MaskLab: Instance Segmentation by Refining Object Detection with Semantic and Direction Features
Authors:
- Liang-Chieh ChenLCLiang-Chieh Chen
- Alexander HermansAHAlexander Hermans
- George PapandreouGPGeorge Papandreou
- Florian SchroffFSFlorian Schroff
- Peng WangPWPeng Wang
- Hartwig AdamHAHartwig Adam
- 1 June 2018
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 4013-4022
Abstract
In this work, we tackle the problem of instance segmentation, the task of simultaneously solving object detection and semantic segmentation. Towards this goal, we present a model, called MaskLab, which produces three outputs: box detection, semantic segmentation, and direction prediction. Building on top of the Faster-RCNN object detector, the predicted boxes provide accurate localization of object instances. Within each region of interest, MaskLab performs foreground/background segmentation by combining semantic and direction prediction. Semantic segmentation assists the model in distinguishing between objects of different semantic classes including background, while the direction prediction, estimating each pixel's direction towards its corresponding center, allows separating instances of the same semantic class. Moreover, we explore the effect of incorporating recent successful methods from both segmentation and detection (e.g., atrous convolution and hypercolumn). Our proposed model is evaluated on the COCO instance segmentation benchmark and shows comparable performance with other state-of-art models.Keywords
All Related Versions
This publication has 45 references indexed in Scilit:
- Modeling local and global deformations in Deep Learning: Epitomic convolution, Multiple Instance Learning, and sliding window detection2015
- Multi-instance object segmentation with occlusion handling2015
- Fully convolutional networks for semantic segmentation2015
- Multiscale Combinatorial Grouping2014
- Scalable Object Detection Using Deep Neural Networks2014
- Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation2014
- Fast image scanning with deep max-pooling convolutional neural networks2013
- Fast High‐Dimensional Filtering Using the Permutohedral LatticeComputer Graphics Forum, 2010
- Gradient-based learning applied to document recognitionProceedings of the IEEE, 1998
- A Real-Time Algorithm for Signal Analysis with the Help of the Wavelet Transform1989