Details
| Original language | English |
|---|---|
| Title of host publication | 2019 International Conference on Computer Vision Workshop, ICCVW 2019 |
| Subtitle of host publication | Proceedings |
| Publisher | Institute of Electrical and Electronics Engineers Inc. |
| Pages | 615-623 |
| Number of pages | 9 |
| ISBN (electronic) | 978-1-7281-5023-9 |
| ISBN (print) | 978-1-7281-5024-6 |
| Publication status | Published - Oct 2019 |
| Event | 2019 IEEE/CVF 17th International Conference on Computer Vision Workshop (ICCVW) - Seoul, Korea, Republic of Duration: 27 Oct 2019 → 28 Oct 2019 |
Publication series
| Name | IEEE International Conference on Computer Vision Workshops |
|---|---|
| ISSN (Print) | 2473-9936 |
| ISSN (electronic) | 2473-9944 |
Abstract
In this paper we propose Structuring AutoEncoders (SAE). SAEs are neural networks which learn a low dimensional representation of data and are additionally enriched with a desired structure in this low dimensional space. While traditional Autoencoders have proven to structure data naturally they fail to discover semantic structure that is hard to recognize in the raw data. The SAE solves the problem by enhancing a traditional Autoencoder using weak supervision to form a structured latent space. In the experiments we demonstrate, that the structured latent space allows for a much more efficient data representation for further tasks such as classification for sparsely labeled data, an efficient choice of data to label, and morphing between classes. To demonstrate the general applicability of our method, we show experiments on the benchmark image datasets MNIST, Fashion-MNIST, DeepFashion2 and on a dataset of 3D human shapes.
Keywords
- Autoencoder, Subspaces
ASJC Scopus subject areas
- Computer Science(all)
- Computer Science Applications
- Computer Science(all)
- Computer Vision and Pattern Recognition
Cite this
- Standard
- Harvard
- Apa
- Vancouver
- BibTeX
- RIS
2019 International Conference on Computer Vision Workshop, ICCVW 2019: Proceedings. Institute of Electrical and Electronics Engineers Inc., 2019. p. 615-623 (IEEE International Conference on Computer Vision Workshops).
Research output: Chapter in book/report/conference proceeding › Conference contribution › Research › peer review
}
TY - GEN
T1 - Structuring autoencoders
AU - Rudolph, Marco
AU - Wandt, Bastian
AU - Rosenhahn, Bodo
N1 - Funding information: This work was funded by the Deutsche Forschungsgemeinschaft (DFG,GermanResearchFoundation) underGermany’s Excellence Strategy within the Cluster of ExcellencePhoenixD (EXC2122).
PY - 2019/10
Y1 - 2019/10
N2 - In this paper we propose Structuring AutoEncoders (SAE). SAEs are neural networks which learn a low dimensional representation of data and are additionally enriched with a desired structure in this low dimensional space. While traditional Autoencoders have proven to structure data naturally they fail to discover semantic structure that is hard to recognize in the raw data. The SAE solves the problem by enhancing a traditional Autoencoder using weak supervision to form a structured latent space. In the experiments we demonstrate, that the structured latent space allows for a much more efficient data representation for further tasks such as classification for sparsely labeled data, an efficient choice of data to label, and morphing between classes. To demonstrate the general applicability of our method, we show experiments on the benchmark image datasets MNIST, Fashion-MNIST, DeepFashion2 and on a dataset of 3D human shapes.
AB - In this paper we propose Structuring AutoEncoders (SAE). SAEs are neural networks which learn a low dimensional representation of data and are additionally enriched with a desired structure in this low dimensional space. While traditional Autoencoders have proven to structure data naturally they fail to discover semantic structure that is hard to recognize in the raw data. The SAE solves the problem by enhancing a traditional Autoencoder using weak supervision to form a structured latent space. In the experiments we demonstrate, that the structured latent space allows for a much more efficient data representation for further tasks such as classification for sparsely labeled data, an efficient choice of data to label, and morphing between classes. To demonstrate the general applicability of our method, we show experiments on the benchmark image datasets MNIST, Fashion-MNIST, DeepFashion2 and on a dataset of 3D human shapes.
KW - Autoencoder
KW - Subspaces
UR - http://www.scopus.com/inward/record.url?scp=85082491587&partnerID=8YFLogxK
U2 - 10.48550/arXiv.1908.02626
DO - 10.48550/arXiv.1908.02626
M3 - Conference contribution
AN - SCOPUS:85082491587
SN - 978-1-7281-5024-6
T3 - IEEE International Conference on Computer Vision Workshops
SP - 615
EP - 623
BT - 2019 International Conference on Computer Vision Workshop, ICCVW 2019
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2019 IEEE/CVF 17th International Conference on Computer Vision Workshop (ICCVW)
Y2 - 27 October 2019 through 28 October 2019
ER -