Abstract | ||
---|---|---|
While convolutional neural networks have gained impressive success recently in solving structured prediction problems such as semantic segmentation, it remains a challenge to differentiate individual object instances in the scene. Instance segmentation is very important in a variety of applications, such as autonomous driving, image captioning, and visual question answering. Techniques that combine large graphical models with low-level vision have been proposed to address this problem, however, we propose an end-to-end recurrent neural network (RNN) architecture with an attention mechanism to model a human-like counting process, and produce detailed instance segmentations. The network is jointly trained to sequentially produce regions of interest as well as a dominant object segmentation within each region. The proposed model achieves competitive results on the CVPPP [27], KITTI [12], and Cityscapes [8] datasets. |
Year | DOI | Venue |
---|---|---|
2017 | 10.1109/CVPR.2017.39 | 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) |
Keywords | Field | DocType |
end-to-end instance segmentation,CVPPP dataset,KITTI dataset,Cityscapes dataset,dominant object segmentation,detailed instance segmentations,attention mechanism,end-to-end recurrent neural network architecture,low-level vision,differentiate individual object instances,semantic segmentation,structured prediction problems,convolutional neural networks | Scale-space segmentation,Convolutional neural network,Segmentation,Computer science,Structured prediction,Segmentation-based object categorization,Recurrent neural network,Image segmentation,Artificial intelligence,Graphical model,Machine learning | Conference |
Volume | Issue | ISSN |
2017 | 1 | 1063-6919 |
ISBN | Citations | PageRank |
978-1-5386-0458-8 | 35 | 1.15 |
References | Authors | |
36 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Mengye Ren | 1 | 265 | 16.34 |
Richard S. Zemel | 2 | 4958 | 425.68 |