TITAN: Bringing The Deep Image Prior to Implicit Representations
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2022
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866917654885826560 |
|---|---|
| author | Luzi, Lorenzo LeJeune, Daniel Siahkoohi, Ali Alemohammad, Sina Saragadam, Vishwanath Babaei, Hossein Liu, Naiming Wang, Zichao Baraniuk, Richard G. |
| author_facet | Luzi, Lorenzo LeJeune, Daniel Siahkoohi, Ali Alemohammad, Sina Saragadam, Vishwanath Babaei, Hossein Liu, Naiming Wang, Zichao Baraniuk, Richard G. |
| contents | We study the interpolation capabilities of implicit neural representations (INRs) of images. In principle, INRs promise a number of advantages, such as continuous derivatives and arbitrary sampling, being freed from the restrictions of a raster grid. However, empirically, INRs have been observed to poorly interpolate between the pixels of the fit image; in other words, they do not inherently possess a suitable prior for natural images. In this paper, we propose to address and improve INRs' interpolation capabilities by explicitly integrating image prior information into the INR architecture via deep decoder, a specific implementation of the deep image prior (DIP). Our method, which we call TITAN, leverages a residual connection from the input which enables integrating the principles of the grid-based DIP into the grid-free INR. Through super-resolution and computed tomography experiments, we demonstrate that our method significantly improves upon classic INRs, thanks to the induced natural image bias. We also find that by constraining the weights to be sparse, image quality and sharpness are enhanced, increasing the Lipschitz constant. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2211_00219 |
| institution | arXiv |
| publishDate | 2022 |
| record_format | arxiv |
| spellingShingle | TITAN: Bringing The Deep Image Prior to Implicit Representations Luzi, Lorenzo LeJeune, Daniel Siahkoohi, Ali Alemohammad, Sina Saragadam, Vishwanath Babaei, Hossein Liu, Naiming Wang, Zichao Baraniuk, Richard G. Image and Video Processing Signal Processing We study the interpolation capabilities of implicit neural representations (INRs) of images. In principle, INRs promise a number of advantages, such as continuous derivatives and arbitrary sampling, being freed from the restrictions of a raster grid. However, empirically, INRs have been observed to poorly interpolate between the pixels of the fit image; in other words, they do not inherently possess a suitable prior for natural images. In this paper, we propose to address and improve INRs' interpolation capabilities by explicitly integrating image prior information into the INR architecture via deep decoder, a specific implementation of the deep image prior (DIP). Our method, which we call TITAN, leverages a residual connection from the input which enables integrating the principles of the grid-based DIP into the grid-free INR. Through super-resolution and computed tomography experiments, we demonstrate that our method significantly improves upon classic INRs, thanks to the induced natural image bias. We also find that by constraining the weights to be sparse, image quality and sharpness are enhanced, increasing the Lipschitz constant. |
| title | TITAN: Bringing The Deep Image Prior to Implicit Representations |
| topic | Image and Video Processing Signal Processing |
| url | https://arxiv.org/abs/2211.00219 |