Naurální radiační pole v elektronové mikroskopii
Thesis title in Czech: | Naurální radiační pole v elektronové mikroskopii |
---|---|
Thesis title in English: | Neural radiance fields in electron microscopy |
Key words: | implicit neural representation|neural field|inverse problems|3D representation |
English key words: | implicit neural representation|neural field|inverse problems|3D representation |
Academic year of topic announcement: | 2023/2024 |
Thesis type: | Bachelor's thesis |
Thesis language: | |
Department: | Department of Software and Computer Science Education (32-KSVI) |
Supervisor: | doc. Ing. Filip Šroubek, Ph.D., DSc. |
Author: | Vladimír Vozár - assigned and confirmed by the Study Dept. |
Date of registration: | 21.02.2024 |
Date of assignment: | 22.02.2024 |
Confirmed by Study dept. on: | 22.02.2024 |
Guidelines |
Neural fields, also called implicit neural representations, have revolutionized the field of inverse problems by representing the unknown function by a neural network with positional encoding and/or periodic activation functions. Particularly appealing results have recently been achieved in the representation of 3D scenes as neural radiance fields - NeRF - for 2D view synthesis. NeRF and many of its recent improvements have been applied to natural scenes, which are often encountered in everyday life. Modern microscopy technology provides acquisition methods that can be used for 3D reconstruction, but neural radiance fields have not yet been considered for the representation of such data. The goal of this thesis is to allow the student to gain a full understanding of neural radiance fields, review the latest approaches, categorize them, implement some of the most promising methods, and apply them to electron microscopy data. |
References |
Ben Mildenhall et al. “NeRF: representing scenes as neural radiance fields for view synthesis.” Commun. ACM 65, 1 (January 2022), 99–106. https://doi.org/10.1145/3503250
Barron, Jonathan T. et al. “Mip-NeRF: A Multiscale Representation for Anti-Aliasing Neural Radiance Fields.” 2021 IEEE/CVF International Conference on Computer Vision (ICCV) (2021): 5835-5844. Matthew Tancik et al. “Block-NeRF: Scalable Large Scene Neural View Synthesis”, 2022, http://arxiv.org/abs/2202.05263v1 Jonathan T. Barron et al. “Zip-NeRF: Anti-Aliased Grid-Based Neural Radiance Fields”, 2023 IEEE/CVF International Conference on Computer Vision (ICCV) (2023) Rundi Wu and Ben Mildenhall “ReconFusion: 3D Reconstruction with Diffusion Priors”, 2023, http://arxiv.org/abs/2312.02981 |