Implementation method of NERF new view angle synthesis model based on self-integration module

The invention discloses an implementation method of an NERF new view angle synthesis model based on a self-integration module. The method comprises the following steps that 1, image data are selected, training data of an open source of an NERF author in googledrive are selected, the training data co...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: LIU ZHENYANG, ZHOU XINLONG, GUO HUANLEI, GUO YIHONG, DING JIAJUN, CHEN TIANXIAO, SUN CHUN, YUAN JUNQING, LIN NINGYU
Format: Patent
Sprache:chi ; eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The invention discloses an implementation method of an NERF new view angle synthesis model based on a self-integration module. The method comprises the following steps that 1, image data are selected, training data of an open source of an NERF author in googledrive are selected, the training data comprise generated Lego data and scene data, and the Lego data and the scene data are stored in two folders respectively; 2, constructing an NERF new view angle synthesis model based on a self-integration module, wherein the NERF new view angle synthesis model comprises the step of obtaining scene representation of NERF; voxel rendering based on scene characterization; and constructing a self-integration module. According to the invention, the self-integration module is added to the NERF neural radiation field, so that the peak signal-to-noise ratio of the output image can be improved, and the image quality is improved. According to the self-integration module provided by the invention, the 3D new view angle reconstr