IEEE Transactions on Pattern Analysis and Machine Intelligence· 2016Q1
Fully Convolutional Networks for Semantic Segmentation
- 11,126citations
- Q1SCImago
- 2016year
Short summary
Fully convolutional networks (FCNs) achieve state-of-the-art semantic segmentation by training end-to-end, pixels-to-pixels, improving mean Intersection over Union (mIoU) by 30% to 67.2% on PASCAL VOC 2012.
AI-generated from the title and abstract; the full text is not read.
Key points
- Fully convolutional networks (FCNs) are introduced for end-to-end semantic segmentation.
- FCNs achieve a 30% relative improvement in mean IoU (to 67.2%) on PASCAL VOC 2012.
- A skip architecture combines deep semantic and shallow appearance information for detailed segmentations.
- Inference is efficient, taking only 0.1 seconds for a typical image.
AI-generated from the title and abstract; the full text is not read.
Abstract
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, improve on the previous best result in semantic segmentation. Our key insight is to build "fully convolutional" networks that take input of arbitrary size and produce correspondingly-sized output with efficient inference and learning. We define and detail the space of fully convolutional networks, explain their application to spatially dense prediction tasks, and draw connections to prior models. We adapt contemporary classification networks (AlexNet, the VGG net, and GoogLeNet) into fully convolutional networks and transfer their learned representations by fine-tuning to the segmentation task. We then define a skip architecture that combines semantic information from a deep, coarse layer with appearance information from a shallow, fine layer to produce accurate and detailed segmentations. Our fully convolutional networks achieve improved segmentation of PASCAL VOC (30% relative improvement to 67.2% mean IU on 2012), NYUDv2, SIFT Flow, and PASCAL-Context, while inference takes one tenth of a second for a typical image.
The authors' abstract, as published at the source. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2016 · DOI ↗
Continue with a free account
Ask the paper: 3 free questions a day about this paper; save it, get its citation, new summaries every day for your field. Takeaways are Premium.
Continue free on the webSign in with Google or Apple; no card needed. You come back to this paper.
On your phone:
Field: Computer Vision and Pattern Recognition
Computer Vision and Pattern RecognitionComputer Science