Sökning: id:"swepub:oai:DiVA.org:umu-204121" >
Equivariance versus...
Equivariance versus augmentation for spherical images
-
- Gerken, Jan, 1991 (författare)
- Gothenburg University,Göteborgs universitet,Institutionen för matematiska vetenskaper, Algebra och geometri,Department of Mathematical Sciences, Algebra and Geometry,Technische Universität Berlin,Chalmers tekniska högskola,Chalmers University of Technology
-
- Carlsson, Oscar, 1996 (författare)
- Gothenburg University,Göteborgs universitet,Institutionen för matematiska vetenskaper, Algebra och geometri,Department of Mathematical Sciences, Algebra and Geometry,Chalmers tekniska högskola,Chalmers University of Technology
-
- Linander, Hampus, 1985 (författare)
- Gothenburg University,Göteborgs universitet,Institutionen för matematiska vetenskaper, Algebra och geometri,Department of Mathematical Sciences, Algebra and Geometry,University of Gothenburg
-
visa fler...
-
- Ohlsson, Fredrik (författare)
- Umeå universitet,Institutionen för matematik och matematisk statistik,Umeå University
-
- Petersson, Christoffer, 1979 (författare)
- Zenseact, Gothenburg, Sweden; Department of Mathematical Sciences, Chalmers University of Technology, Gothenburg, Sweden,Chalmers tekniska högskola,Chalmers University of Technology
-
- Persson, Daniel, 1978 (författare)
- Gothenburg University,Göteborgs universitet,Institutionen för matematiska vetenskaper, Algebra och geometri,Department of Mathematical Sciences, Algebra and Geometry,Chalmers tekniska högskola,Chalmers University of Technology
-
visa färre...
-
(creator_code:org_t)
- 2022
- 2022
- Engelska.
-
Ingår i: Proceedings of Machine Learning Resaerch. ; , s. 7404-7421
- Relaterad länk:
-
https://proceedings....
-
visa fler...
-
https://urn.kb.se/re...
-
https://gup.ub.gu.se...
-
https://research.cha...
-
visa färre...
Abstract
Ämnesord
Stäng
- We analyze the role of rotational equivariance in convolutional neural networks (CNNs) applied to spherical images. We compare the performance of the group equivariant networks known as S2CNNs and standard non-equivariant CNNs trained with an increasing amount of data augmentation. The chosen architectures can be considered baseline references for the respective design paradigms. Our models are trained and evaluated on single or multiple items from the MNIST- or FashionMNIST dataset projected onto the sphere. For the task of image classification, which is inherently rotationally invariant, we find that by considerably increasing the amount of data augmentation and the size of the networks, it is possible for the standard CNNs to reach at least the same performance as the equivariant network. In contrast, for the inherently equivariant task of semantic segmentation, the non-equivariant networks are consistently outperformed by the equivariant networks with significantly fewer parameters. We also analyze and compare the inference latency and training times of the different networks, enabling detailed tradeoff considerations between equivariant architectures and data augmentation for practical problems.
Ämnesord
- NATURVETENSKAP -- Matematik -- Geometri (hsv//swe)
- NATURAL SCIENCES -- Mathematics -- Geometry (hsv//eng)
- NATURVETENSKAP -- Matematik -- Beräkningsmatematik (hsv//swe)
- NATURAL SCIENCES -- Mathematics -- Computational Mathematics (hsv//eng)
- NATURVETENSKAP -- Data- och informationsvetenskap -- Datorseende och robotik (hsv//swe)
- NATURAL SCIENCES -- Computer and Information Sciences -- Computer Vision and Robotics (hsv//eng)
- NATURVETENSKAP -- Data- och informationsvetenskap -- Annan data- och informationsvetenskap (hsv//swe)
- NATURAL SCIENCES -- Computer and Information Sciences -- Other Computer and Information Science (hsv//eng)
- NATURVETENSKAP -- Data- och informationsvetenskap -- Datorteknik (hsv//swe)
- NATURAL SCIENCES -- Computer and Information Sciences -- Computer Engineering (hsv//eng)
- TEKNIK OCH TEKNOLOGIER -- Elektroteknik och elektronik -- Kommunikationssystem (hsv//swe)
- ENGINEERING AND TECHNOLOGY -- Electrical Engineering, Electronic Engineering, Information Engineering -- Communication Systems (hsv//eng)
- NATURVETENSKAP -- Data- och informationsvetenskap -- Bioinformatik (hsv//swe)
- NATURAL SCIENCES -- Computer and Information Sciences -- Bioinformatics (hsv//eng)
- NATURVETENSKAP -- Matematik -- Matematisk analys (hsv//swe)
- NATURAL SCIENCES -- Mathematics -- Mathematical Analysis (hsv//eng)
Nyckelord
- Mathematics
- matematik
Publikations- och innehållstyp
- ref (ämneskategori)
- kon (ämneskategori)
Hitta via bibliotek
Till lärosätets databas