Please use this identifier to cite or link to this item:
doi:10.22028/D291-39703
Title: | Deep spatial and tonal data optimisation for homogeneous diffusion inpainting |
Author(s): | Peter, Pascal Schrader, Karl Alt, Tobias Weickert, Joachim |
Language: | English |
Title: | Pattern Analysis and Applications |
Publisher/Platform: | Springer Nature |
Year of Publication: | 2023 |
Free key words: | Image inpainting Difusion Partial diferential equations Data optimisation Deep learning |
DDC notations: | 004 Computer science, internet |
Publikation type: | Journal Article |
Abstract: | Difusion-based inpainting can reconstruct missing image areas with high quality from sparse data, provided that their location and their values are well optimised. This is particularly useful for applications such as image compression, where the original image is known. Selecting the known data constitutes a challenging optimisation problem, that has so far been only investigated with model-based approaches. So far, these methods require a choice between either high quality or high speed since qualitatively convincing algorithms rely on many time-consuming inpaintings. We propose the frst neural network architecture that allows fast optimisation of pixel positions and pixel values for homogeneous difusion inpainting. During training, we combine two optimisation networks with a neural network-based surrogate solver for difusion inpainting. This novel concept allows us to perform backpropagation based on inpainting results that approximate the solution of the inpainting equation. Without the need for a single inpainting during test time, our deep optimisation accelerates data selection by more than four orders of magnitude compared to common model-based approaches. This provides real-time performance with high quality results. |
DOI of the first publication: | 10.1007/s10044-023-01162-y |
URL of the first publication: | https://link.springer.com/article/10.1007/s10044-023-01162-y |
Link to this record: | urn:nbn:de:bsz:291--ds-397032 hdl:20.500.11880/35773 http://dx.doi.org/10.22028/D291-39703 |
ISSN: | 1433-755X 1433-7541 |
Date of registration: | 8-May-2023 |
Faculty: | MI - Fakultät für Mathematik und Informatik |
Department: | MI - Informatik |
Professorship: | MI - Prof. Dr. Joachim Weickert |
Collections: | SciDok - Der Wissenschaftsserver der Universität des Saarlandes |
Files for this record:
File | Description | Size | Format | |
---|---|---|---|---|
s10044-023-01162-y.pdf | 3,97 MB | Adobe PDF | View/Open |
This item is licensed under a Creative Commons License