Abstract

In this paper, we propose Sequential Grouping Networks (SGN) to tackle the problem of object instance segmentation. SGNs employ a sequence of neural networks, each solving a sub-grouping problem of increasing semantic complexity in order to gradually compose objects out of pixels. In particular, the first network aims to group pixels along each image row and column by predicting horizontal and vertical object breakpoints. These breakpoints are then used to create line segments. By exploiting two-directional information, the second network groups horizontal and vertical lines into connected components. Finally, the third network groups the connected components into object instances. Our experiments show that our SGN significantly outperforms state-of-the-art approaches in both, the Cityscapes dataset as well as PASCAL VOC. © 2017 IEEE.

Keywords

Pascal (unit)Computer sciencePixelSegmentationObject (grammar)Image segmentationArtificial intelligencePattern recognition (psychology)Artificial neural networkLine segmentSequence (biology)Line (geometry)Computer visionMathematics

Affiliated Institutions

Related Publications

Normalized cuts and image segmentation

We propose a novel approach for solving the perceptual grouping problem in vision. Rather than focusing on local features and their consistencies in the image data, our approach...

2000 IEEE Transactions on Pattern Analysis... 15440 citations

Publication Info

Year
2017
Type
article
Citations
295
Access
Closed

External Links

Social Impact

Social media, news, blog, policy document mentions

Citation Metrics

295
OpenAlex

Cite This

Shu Liu, Jiaya Jia, Sanja Fidler et al. (2017). SGN: Sequential Grouping Networks for Instance Segmentation. . https://doi.org/10.1109/iccv.2017.378

Identifiers

DOI
10.1109/iccv.2017.378