Wirth, T. ; Rak, A. ; Knauthe, V. ; Fellner, D. W. (2024)
A Post Processing Technique to Automatically Remove Floater Artifacts in Neural Radiance Fields.
In: Computer Graphics Forum, 2023, 42 (7)
doi: 10.26083/tuprints-00027236
Artikel, Zweitveröffentlichung, Verlagsversion
 | Es ist eine neuere Version dieses Eintrags verfügbar. |
Kurzbeschreibung (Abstract)
Neural Radiance Fields have revolutionized Novel View Synthesis by providing impressive levels of realism. However, in most in‐the‐wild scenes they suffer from floater artifacts that occur due to sparse input images or strong view‐dependent effects. We propose an approach that uses neighborhood based clustering and a consistency metric on NeRF models trained on different scene scales to identify regions that contain floater artifacts based on Instant‐NGPs multiscale occupancy grids. These occupancy grids contain the position of relevant optical densities in the scene. By pruning the regions that we identified as containing floater artifacts, they are omitted during the rendering process, leading to higher quality resulting images. Our approach has no negative runtime implications for the rendering process and does not require retraining of the underlying Multi Layer Perceptron. We show on a qualitative base, that our approach is suited to remove floater artifacts while preserving most of the scenes relevant geometry. Furthermore, we conduct a comparison to state‐of‐the‐art techniques on the Nerfbusters dataset, that was created with measuring the implications of floater artifacts in mind. This comparison shows, that our method outperforms currently available techniques. Our approach does not require additional user input, but can be be used in an interactive manner. In general, the presented approach is applicable to every architecture that uses an explicit representation of a scene's occupancy distribution to accelerate the rendering process.
| Typ des Eintrags: | Artikel |
|---|---|
| Erschienen: | 2024 |
| Autor(en): | Wirth, T. ; Rak, A. ; Knauthe, V. ; Fellner, D. W. |
| Art des Eintrags: | Zweitveröffentlichung |
| Titel: | A Post Processing Technique to Automatically Remove Floater Artifacts in Neural Radiance Fields |
| Sprache: | Englisch |
| Publikationsjahr: | 27 Mai 2024 |
| Ort: | Darmstadt |
| Publikationsdatum der Erstveröffentlichung: | Oktober 2023 |
| Ort der Erstveröffentlichung: | Oxford |
| Verlag: | Wiley-Blackwell |
| Titel der Zeitschrift, Zeitung oder Schriftenreihe: | Computer Graphics Forum |
| Jahrgang/Volume einer Zeitschrift: | 42 |
| (Heft-)Nummer: | 7 |
| Kollation: | 12 Seiten |
| DOI: | 10.26083/tuprints-00027236 |
| URL / URN: | https://tuprints.ulb.tu-darmstadt.de/27236 |
| Zugehörige Links: | |
| Herkunft: | Zweitveröffentlichung DeepGreen |
| Kurzbeschreibung (Abstract): | Neural Radiance Fields have revolutionized Novel View Synthesis by providing impressive levels of realism. However, in most in‐the‐wild scenes they suffer from floater artifacts that occur due to sparse input images or strong view‐dependent effects. We propose an approach that uses neighborhood based clustering and a consistency metric on NeRF models trained on different scene scales to identify regions that contain floater artifacts based on Instant‐NGPs multiscale occupancy grids. These occupancy grids contain the position of relevant optical densities in the scene. By pruning the regions that we identified as containing floater artifacts, they are omitted during the rendering process, leading to higher quality resulting images. Our approach has no negative runtime implications for the rendering process and does not require retraining of the underlying Multi Layer Perceptron. We show on a qualitative base, that our approach is suited to remove floater artifacts while preserving most of the scenes relevant geometry. Furthermore, we conduct a comparison to state‐of‐the‐art techniques on the Nerfbusters dataset, that was created with measuring the implications of floater artifacts in mind. This comparison shows, that our method outperforms currently available techniques. Our approach does not require additional user input, but can be be used in an interactive manner. In general, the presented approach is applicable to every architecture that uses an explicit representation of a scene's occupancy distribution to accelerate the rendering process. |
| ID-Nummer: | Artikel-ID: e14977 |
| Status: | Verlagsversion |
| URN: | urn:nbn:de:tuda-tuprints-272362 |
| Sachgruppe der Dewey Dezimalklassifikatin (DDC): | 000 Allgemeines, Informatik, Informationswissenschaft > 004 Informatik |
| Fachbereich(e)/-gebiet(e): | 20 Fachbereich Informatik 20 Fachbereich Informatik > Graphisch-Interaktive Systeme 20 Fachbereich Informatik > Fraunhofer IGD |
| Hinterlegungsdatum: | 27 Mai 2024 12:55 |
| Letzte Änderung: | 03 Jun 2024 11:56 |
| PPN: | |
| Export: | |
| Suche nach Titel in: | TUfind oder in Google |
Verfügbare Versionen dieses Eintrags
- A Post Processing Technique to Automatically Remove Floater Artifacts in Neural Radiance Fields. (deposited 27 Mai 2024 12:55) [Gegenwärtig angezeigt]
 |
Frage zum Eintrag |
Optionen (nur für Redakteure)
 |
Redaktionelle Details anzeigen |
Drucken
Impressum