A Spitting Image: Modular Superpixel Tokenization in Vision Transformers

Autor: Aasan, Marius, Kolbjørnsen, Odd, Solberg, Anne Schistad, Rivera, Adín Ramirez

Vision Transformer (ViT) architectures traditionally employ a grid-based approach to tokenization independent of the semantic content of an image. We propose a modular superpixel tokenization strategy which decouples tokenization and feature extracti

Externí odkaz: http://arxiv.org/abs/2408.07680

Multi-task neural networks by learned contextual inputs

Autor: Sandnes, Anders T., Grimstad, Bjarne, Kolbjørnsen, Odd

This paper explores learned-context neural networks. It is a multi-task learning architecture based on a fully shared neural network and an augmented input vector containing trainable task parameters. The architecture is interesting due to its powerf

Externí odkaz: http://arxiv.org/abs/2303.00788

Multi-task learning for virtual flow metering

Autor: Sandnes, Anders T., Grimstad, Bjarne, Kolbjørnsen, Odd

Virtual flow metering (VFM) is a cost-effective and non-intrusive technology for inferring multiphase flow rates in petroleum assets. Inferences about flow rates are fundamental to decision support systems that operators extensively rely on. Data-dri

Externí odkaz: http://arxiv.org/abs/2103.08713

Nonlinear Topics in the Bayesian Approach to Inverse Problems with Applications to Seismic Inversion

Autor: Kolbjørnsen, Odd

Paper I considers piecewise affine inverse problems. This is a large group of nonlinear inverse problems. Problems that obey certain variational structures are of this type. In inverse problems it is frequently such that some features are well determ

Externí odkaz: http://urn.kb.se/resolve?urn=urn:nbn:no:ntnu:diva-489

Cross-streamer wavefield interpolation using deep convolutional networks

Autor: Greiner, Thomas Larsen, Kolbjørnsen, Odd, Lie, Jan Erik, Harris Nilsen, Espen, Kjeldsrud Evensen, Andreas, Gelius, Leiv-J.

Publikováno v: Greiner, Thomas Larsen Kolbjørnsen, Odd Lie, Jan Erik Harris Nilsen, Espen Kjeldsrud Evensen, Andreas Gelius, Leiv-J. . Cross-streamer wavefield interpolation using deep convolutional networks. SEG Technical Program Expanded Abstracts 2019. 2019. USA: SEG

Externí odkaz: http://hdl.handle.net/10852/77027
https://www.duo.uio.no/bitstream/handle/10852/77027/1/2019-Cross_Streamer_Wavefield_Interpolation.pdf