1. Identity statement | |
Reference Type | Conference Paper (Conference Proceedings) |
Site | sibgrapi.sid.inpe.br |
Identifier | 8JMKD3MGPEW34M/47N4R68 |
Repository | sid.inpe.br/sibgrapi/2022/09.30.02.43 |
Last Update | 2022:09.30.02.43.41 (UTC) jpklock@ufmg.br |
Metadata Repository | sid.inpe.br/sibgrapi/2022/09.30.02.43.41 |
Metadata Last Update | 2023:05.23.04.20.43 (UTC) administrator |
Citation Key | FerreiraPintCast:2022:SeScGe |
Title | Weaklier Supervised: Semi-automatic Scribble Generation Applied to Semantic Segmentation ![](http://sibgrapi.sid.inpe.br/col/dpi.inpe.br/banon/2000/01.23.20.24/doc/externalLink.gif) |
Format | On-line |
Year | 2022 |
Access Date | 2024, July 27 |
Number of Files | 1 |
Size | 6896 KiB |
|
2. Context | |
Author | 1 Ferreira, Joćo Pedro Klock 2 Pinto, Joćo Paulo Lara 3 Castro, Cristiano Leite |
Affiliation | 1 Graduate Program in Electrical Engineering - Universidade Federal de Minas Gerais 2 Graduate Program in Electrical Engineering - Universidade Federal de Minas Gerais 3 Graduate Program in Electrical Engineering - Universidade Federal de Minas Gerais |
e-Mail Address | jpklock@ufmg.br |
Conference Name | Conference on Graphics, Patterns and Images, 35 (SIBGRAPI) |
Conference Location | Natal, RN |
Date | 24-27 Oct. 2022 |
Book Title | Proceedings |
Tertiary Type | Work in Progress |
History (UTC) | 2022-09-30 02:43:41 :: jpklock@ufmg.br -> administrator :: 2023-05-23 04:20:43 :: administrator -> :: 2022 |
|
3. Content and structure | |
Is the master or a copy? | is the master |
Content Stage | completed |
Transferable | 1 |
Keywords | Weak-supervision semi-supervision scribbles semantic segmentation remote sensing |
Abstract | With many applications regarding semantic segmentation arising, along with the advent of the Deep Semantic Segmentation Networks, the need for large labeled datasets has also largely increased. But labeling thousands of images can be very expensive and time-consuming. Approaches such as weak and semi supervision try do deal with this problem, but the first cannot deal with large datasets and the latter is hard to deal with semantic segmentation. Therefore, in this work we propose a combination of both to create a novel pipeline of weak supervision, with focus in satellite imagery, capable of dealing with large datasets. We propose a pipeline to automatically generate scribbles in images, requiring that the user only label 10% of the images in a given dataset, while a classifier deal with the remaining images. Along with that, we also propose a simple semantic segmentation pipeline, that uses only images with scribbles to train a network. Results show that performance is lower, but similar to a fully supervised pipeline. |
Arrangement | urlib.net > SDLA > Fonds > SIBGRAPI 2022 > Weaklier Supervised: Semi-automatic... |
doc Directory Content | access |
source Directory Content | there are no files |
agreement Directory Content | |
|
4. Conditions of access and use | |
data URL | http://urlib.net/ibi/8JMKD3MGPEW34M/47N4R68 |
zipped data URL | http://urlib.net/zip/8JMKD3MGPEW34M/47N4R68 |
Language | en |
Target File | Ferreira-5-no-copyright.pdf |
User Group | jpklock@ufmg.br |
Visibility | shown |
|
5. Allied materials | |
Mirror Repository | sid.inpe.br/banon/2001/03.30.15.38.24 |
Next Higher Units | 8JMKD3MGPEW34M/495MHJ8 |
Citing Item List | sid.inpe.br/sibgrapi/2023/05.19.12.10 19 sid.inpe.br/banon/2001/03.30.15.38.24 8 |
Host Collection | sid.inpe.br/banon/2001/03.30.15.38 |
|
6. Notes | |
Empty Fields | archivingpolicy archivist area callnumber contenttype copyholder copyright creatorhistory descriptionlevel dissemination documentstage doi edition editor electronicmailaddress group holdercode isbn issn label lineage mark nextedition notes numberofvolumes orcid organization pages parameterlist parentrepositories previousedition previouslowerunit progress project publisher publisheraddress readergroup readpermission resumeid rightsholder schedulinginformation secondarydate secondarykey secondarymark secondarytype serieseditor session shorttitle sponsor subject tertiarymark type url versiontype volume |
|