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.
Keywords
Computer scienceArtificial intelligenceSegmentationClosed captioningEnd-to-end principleConvolutional neural networkObject (grammar)Recurrent neural networkProcess (computing)Image segmentationMachine learningObject detectionSemantics (computer science)Variety (cybernetics)Computer visionArtificial neural networkPattern recognition (psychology)Image (mathematics)
Affiliated Institutions
Related Publications
Fully convolutional networks for semantic segmentation
Convolutional networks are powerful visual models that yield hierarchies of features. We show that convolutional networks by themselves, trained end-to-end, pixels-to-pixels, ex...
FCOS: Fully Convolutional One-Stage Object Detection
We propose a fully convolutional one-stage object detector (FCOS) to solve object detection in a per-pixel prediction fashion, analogue to semantic segmentation. Almost all stat...
CutMix: Regularization Strategy to Train Strong Classifiers With Localizable Features
Regional dropout strategies have been proposed to enhance performance of convolutional neural network classifiers. They have proved to be effective for guiding the model to atte...
Path Aggregation Network for Instance Segmentation
The way that information propagates in neural networks is of great importance. In this paper, we propose Path Aggregation Network (PANet) aiming at boosting information flow in ...
Publication Info
- Year
- 2017
- Type
- preprint
- Pages
- 293-301
- Citations
- 314
- Access
- Closed
External Links
Social Impact
Altmetric
PlumX Metrics
Social media, news, blog, policy document mentions
Citation Metrics
314
OpenAlex
Cite This
Mengye Ren,
Richard S. Zemel
(2017).
End-to-End Instance Segmentation with Recurrent Attention.
, 293-301.
https://doi.org/10.1109/cvpr.2017.39
Identifiers
- DOI
- 10.1109/cvpr.2017.39