Multi-scale context aggregation by dilated convolutions. Yu, F. & Koltun, V. 4th International Conference on Learning Representations, ICLR 2016 - Conference Track Proceedings, 2016. ![Multi-scale context aggregation by dilated convolutions [pdf]](https://bibbase.org/img/filetypes/pdf.svg "pdf")
Paper abstract bibtex State-of-the-art models for semantic segmentation are based on adaptations of convolutional networks that had originally been designed for image classification. However, dense prediction problems such as semantic segmentation are structurally different from image classification. In this work, we develop a new convolutional network module that is specifically designed for dense prediction. The presented module uses dilated convolutions to systematically aggregate multi-scale contextual information without losing resolution. The architecture is based on the fact that dilated convolutions support exponential expansion of the receptive field without loss of resolution or coverage. We show that the presented context module increases the accuracy of state-of-the-art semantic segmentation systems. In addition, we examine the adaptation of image classification networks to dense prediction and show that simplifying the adapted network can increase accuracy.
@article{
title = {Multi-scale context aggregation by dilated convolutions},
type = {article},
year = {2016},
id = {a6d94979-ac3c-3ef6-a79f-9a6d5602cf15},
created = {2021-11-01T10:14:38.777Z},
file_attached = {true},
profile_id = {235249c2-3ed4-314a-b309-b1ea0330f5d9},
group_id = {1ff583c0-be37-34fa-9c04-73c69437d354},
last_modified = {2022-03-28T09:45:09.218Z},
read = {false},
starred = {false},
authored = {false},
confirmed = {true},
hidden = {false},
citation_key = {Yu2016},
folder_uuids = {cd02f564-0123-4236-a320-b339927f085a},
private_publication = {false},
abstract = {State-of-the-art models for semantic segmentation are based on adaptations of convolutional networks that had originally been designed for image classification. However, dense prediction problems such as semantic segmentation are structurally different from image classification. In this work, we develop a new convolutional network module that is specifically designed for dense prediction. The presented module uses dilated convolutions to systematically aggregate multi-scale contextual information without losing resolution. The architecture is based on the fact that dilated convolutions support exponential expansion of the receptive field without loss of resolution or coverage. We show that the presented context module increases the accuracy of state-of-the-art semantic segmentation systems. In addition, we examine the adaptation of image classification networks to dense prediction and show that simplifying the adapted network can increase accuracy.},
bibtype = {article},
author = {Yu, Fisher and Koltun, Vladlen},
journal = {4th International Conference on Learning Representations, ICLR 2016 - Conference Track Proceedings}
}
Downloads: 0
{"_id":"C5M3Jca3D69g4Cm43","bibbaseid":"yu-koltun-multiscalecontextaggregationbydilatedconvolutions-2016","author_short":["Yu, F.","Koltun, V."],"bibdata":{"title":"Multi-scale context aggregation by dilated convolutions","type":"article","year":"2016","id":"a6d94979-ac3c-3ef6-a79f-9a6d5602cf15","created":"2021-11-01T10:14:38.777Z","file_attached":"true","profile_id":"235249c2-3ed4-314a-b309-b1ea0330f5d9","group_id":"1ff583c0-be37-34fa-9c04-73c69437d354","last_modified":"2022-03-28T09:45:09.218Z","read":false,"starred":false,"authored":false,"confirmed":"true","hidden":false,"citation_key":"Yu2016","folder_uuids":"cd02f564-0123-4236-a320-b339927f085a","private_publication":false,"abstract":"State-of-the-art models for semantic segmentation are based on adaptations of convolutional networks that had originally been designed for image classification. However, dense prediction problems such as semantic segmentation are structurally different from image classification. In this work, we develop a new convolutional network module that is specifically designed for dense prediction. The presented module uses dilated convolutions to systematically aggregate multi-scale contextual information without losing resolution. The architecture is based on the fact that dilated convolutions support exponential expansion of the receptive field without loss of resolution or coverage. We show that the presented context module increases the accuracy of state-of-the-art semantic segmentation systems. In addition, we examine the adaptation of image classification networks to dense prediction and show that simplifying the adapted network can increase accuracy.","bibtype":"article","author":"Yu, Fisher and Koltun, Vladlen","journal":"4th International Conference on Learning Representations, ICLR 2016 - Conference Track Proceedings","bibtex":"@article{\n title = {Multi-scale context aggregation by dilated convolutions},\n type = {article},\n year = {2016},\n id = {a6d94979-ac3c-3ef6-a79f-9a6d5602cf15},\n created = {2021-11-01T10:14:38.777Z},\n file_attached = {true},\n profile_id = {235249c2-3ed4-314a-b309-b1ea0330f5d9},\n group_id = {1ff583c0-be37-34fa-9c04-73c69437d354},\n last_modified = {2022-03-28T09:45:09.218Z},\n read = {false},\n starred = {false},\n authored = {false},\n confirmed = {true},\n hidden = {false},\n citation_key = {Yu2016},\n folder_uuids = {cd02f564-0123-4236-a320-b339927f085a},\n private_publication = {false},\n abstract = {State-of-the-art models for semantic segmentation are based on adaptations of convolutional networks that had originally been designed for image classification. However, dense prediction problems such as semantic segmentation are structurally different from image classification. In this work, we develop a new convolutional network module that is specifically designed for dense prediction. The presented module uses dilated convolutions to systematically aggregate multi-scale contextual information without losing resolution. The architecture is based on the fact that dilated convolutions support exponential expansion of the receptive field without loss of resolution or coverage. We show that the presented context module increases the accuracy of state-of-the-art semantic segmentation systems. In addition, we examine the adaptation of image classification networks to dense prediction and show that simplifying the adapted network can increase accuracy.},\n bibtype = {article},\n author = {Yu, Fisher and Koltun, Vladlen},\n journal = {4th International Conference on Learning Representations, ICLR 2016 - Conference Track Proceedings}\n}","author_short":["Yu, F.","Koltun, V."],"urls":{"Paper":"https://bibbase.org/service/mendeley/bfbbf840-4c42-3914-a463-19024f50b30c/file/288e3e62-f287-0c27-277f-affc6e88777a/Yu_and_Koltun___2016___Multi_Scale_Context_Aggregation_by_Dilated_Convolu.pdf.pdf"},"biburl":"https://bibbase.org/service/mendeley/bfbbf840-4c42-3914-a463-19024f50b30c","bibbaseid":"yu-koltun-multiscalecontextaggregationbydilatedconvolutions-2016","role":"author","metadata":{"authorlinks":{}}},"bibtype":"article","biburl":"https://bibbase.org/service/mendeley/bfbbf840-4c42-3914-a463-19024f50b30c","dataSources":["CmHEoydhafhbkXXt5","ya2CyA73rpZseyrZ8","2252seNhipfTmjEBQ"],"keywords":[],"search_terms":["multi","scale","context","aggregation","dilated","convolutions","yu","koltun"],"title":"Multi-scale context aggregation by dilated convolutions","year":2016}