EDEN: Deep Feature Distribution Pooling for Saimaa Ringed Seals Pattern Matching
In this paper, pelage pattern matching is considered to solve the individual re-identification of the Saimaa ringed seals. Animal reidentification, together with the access to a large amount of image material through camera traps and crowd-sourcing, provides novel possibilities for animal monitoring...
Main Authors: | , , , , |
---|---|
Other Authors: | , , |
Format: | Conference Object |
Language: | English |
Published: |
Springer, Cham
2023
|
Subjects: | |
Online Access: | https://lutpub.lut.fi/handle/10024/166589 |
id |
ftlappeenranta:oai:lutpub.lut.fi:10024/166589 |
---|---|
record_format |
openpolar |
spelling |
ftlappeenranta:oai:lutpub.lut.fi:10024/166589 2024-02-11T10:08:16+01:00 EDEN: Deep Feature Distribution Pooling for Saimaa Ringed Seals Pattern Matching Chelak, Ilia Nepovinnykh, Ekaterina Eerola, Tuomas Kälviäinen, Heikki Belykh, Igor Lappeenrannan-Lahden teknillinen yliopisto LUT Lappeenranta-Lahti University of Technology LUT fi=School of Engineering Science|en=School of Engineering Science| 2023-01-21 10 fulltext 141-150 https://lutpub.lut.fi/handle/10024/166589 eng eng Springer, Cham Cyber-Physical Systems and Control II. CPS&C 2021 460 2367-3370 2367-3389 Lecture Notes in Networks and Systems Cyber-Physical Systems and Control II 978-3-031-20875-1 https://link.springer.com/chapter/10.1007/978-3-031-20875-1_13 https://doi.org/10.1007/978-3-031-20875-1_13 Chelak, I., Nepovinnykh, E., Eerola, T., Kälviäinen, H., Belykh, I. (2023). EDEN: Deep Feature Distribution Pooling for Saimaa Ringed Seals Pattern Matching. In: Arseniev, D.G., Aouf, N. (eds) Cyber-Physical Systems and Control II. CPS&C 2021. Lecture Notes in Networks and Systems, vol 460. Springer, Cham. https://doi.org/10.1007/978-3-031-20875-1_13 https://lutpub.lut.fi/handle/10024/166589 URN:NBN:fi-fe20231123148683 fi=Kaikki oikeudet pidätetään.|en=All rights reserved.| openAccess © 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG Pattern matching Global pooling Animal biometrics Saimaa ringed seals conferenceObject acceptedVersion 2023 ftlappeenranta https://doi.org/10.1007/978-3-031-20875-1_13 2024-01-25T00:04:43Z In this paper, pelage pattern matching is considered to solve the individual re-identification of the Saimaa ringed seals. Animal reidentification, together with the access to a large amount of image material through camera traps and crowd-sourcing, provides novel possibilities for animal monitoring and conservation. Image retrieval techniques, such as global pooling, can be used to solve the individual re-identification. However, current global pooling methods incorporate only value distribution of features, losing spatial information. To overcome the problem, we propose a novel pooling approach that allows aggregating the local pattern features to get a fixed size embedding vector that incorporates global features by taking into account their spatial distribution. This is obtained by eigen decomposition of covariances computed for probability mass functions representing feature maps. Embedding vectors can then be used to find the best match in the database of known individuals allowing animal re-identification. The results show that the proposed pooling technique outperforms the existing methods on the challenging Saimaa ringed seal image data. Post-print / Final draft Conference Object ringed seal LUTPub (LUT University) 141 150 |
institution |
Open Polar |
collection |
LUTPub (LUT University) |
op_collection_id |
ftlappeenranta |
language |
English |
topic |
Pattern matching Global pooling Animal biometrics Saimaa ringed seals |
spellingShingle |
Pattern matching Global pooling Animal biometrics Saimaa ringed seals Chelak, Ilia Nepovinnykh, Ekaterina Eerola, Tuomas Kälviäinen, Heikki Belykh, Igor EDEN: Deep Feature Distribution Pooling for Saimaa Ringed Seals Pattern Matching |
topic_facet |
Pattern matching Global pooling Animal biometrics Saimaa ringed seals |
description |
In this paper, pelage pattern matching is considered to solve the individual re-identification of the Saimaa ringed seals. Animal reidentification, together with the access to a large amount of image material through camera traps and crowd-sourcing, provides novel possibilities for animal monitoring and conservation. Image retrieval techniques, such as global pooling, can be used to solve the individual re-identification. However, current global pooling methods incorporate only value distribution of features, losing spatial information. To overcome the problem, we propose a novel pooling approach that allows aggregating the local pattern features to get a fixed size embedding vector that incorporates global features by taking into account their spatial distribution. This is obtained by eigen decomposition of covariances computed for probability mass functions representing feature maps. Embedding vectors can then be used to find the best match in the database of known individuals allowing animal re-identification. The results show that the proposed pooling technique outperforms the existing methods on the challenging Saimaa ringed seal image data. Post-print / Final draft |
author2 |
Lappeenrannan-Lahden teknillinen yliopisto LUT Lappeenranta-Lahti University of Technology LUT fi=School of Engineering Science|en=School of Engineering Science| |
format |
Conference Object |
author |
Chelak, Ilia Nepovinnykh, Ekaterina Eerola, Tuomas Kälviäinen, Heikki Belykh, Igor |
author_facet |
Chelak, Ilia Nepovinnykh, Ekaterina Eerola, Tuomas Kälviäinen, Heikki Belykh, Igor |
author_sort |
Chelak, Ilia |
title |
EDEN: Deep Feature Distribution Pooling for Saimaa Ringed Seals Pattern Matching |
title_short |
EDEN: Deep Feature Distribution Pooling for Saimaa Ringed Seals Pattern Matching |
title_full |
EDEN: Deep Feature Distribution Pooling for Saimaa Ringed Seals Pattern Matching |
title_fullStr |
EDEN: Deep Feature Distribution Pooling for Saimaa Ringed Seals Pattern Matching |
title_full_unstemmed |
EDEN: Deep Feature Distribution Pooling for Saimaa Ringed Seals Pattern Matching |
title_sort |
eden: deep feature distribution pooling for saimaa ringed seals pattern matching |
publisher |
Springer, Cham |
publishDate |
2023 |
url |
https://lutpub.lut.fi/handle/10024/166589 |
genre |
ringed seal |
genre_facet |
ringed seal |
op_relation |
Cyber-Physical Systems and Control II. CPS&C 2021 460 2367-3370 2367-3389 Lecture Notes in Networks and Systems Cyber-Physical Systems and Control II 978-3-031-20875-1 https://link.springer.com/chapter/10.1007/978-3-031-20875-1_13 https://doi.org/10.1007/978-3-031-20875-1_13 Chelak, I., Nepovinnykh, E., Eerola, T., Kälviäinen, H., Belykh, I. (2023). EDEN: Deep Feature Distribution Pooling for Saimaa Ringed Seals Pattern Matching. In: Arseniev, D.G., Aouf, N. (eds) Cyber-Physical Systems and Control II. CPS&C 2021. Lecture Notes in Networks and Systems, vol 460. Springer, Cham. https://doi.org/10.1007/978-3-031-20875-1_13 https://lutpub.lut.fi/handle/10024/166589 URN:NBN:fi-fe20231123148683 |
op_rights |
fi=Kaikki oikeudet pidätetään.|en=All rights reserved.| openAccess © 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG |
op_doi |
https://doi.org/10.1007/978-3-031-20875-1_13 |
container_start_page |
141 |
op_container_end_page |
150 |
_version_ |
1790607315653623808 |