Novel View Synthesis of Defocused Blur Scenes Based on Neural Radiance Fields

In recent years, neural radiance fields have been widely used in the field of computer graphics due to their excellent reconstruction quality. However, the shooting process in the wild environment is often affected by various internal and external factors, resulting in blurry images. To address the...

Full description

Saved in:
Bibliographic Details
Main Authors: Zhaoji Lin, Yuxin Zheng, Li Yao
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/11017661/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849472035638476800
author Zhaoji Lin
Yuxin Zheng
Li Yao
author_facet Zhaoji Lin
Yuxin Zheng
Li Yao
author_sort Zhaoji Lin
collection DOAJ
description In recent years, neural radiance fields have been widely used in the field of computer graphics due to their excellent reconstruction quality. However, the shooting process in the wild environment is often affected by various internal and external factors, resulting in blurry images. To address the problem of defocus blur in the real world leading to a decrease in the reconstruction quality of neural radiance fields, this paper proposes a new deblurred radiance field and designs a rigid blur kernel based on the depth features of the image frame to model the rigid transformation of light and the weights of the coarse components of color. For the problem of similar two-dimensional coordinates restricting the model to distinguish scene details in the non-focal plane background, a fine sampling weight using multiscale depth feature fusion is further proposed, and a staged optimization strategy is designed. Experimental results show that compared with the state-of-the-art methods, the method proposed in this paper can better recover scene details and generate high-quality images for defocused blur scenes.
format Article
id doaj-art-8faa06e1e3064f6f989cd8dddee70c0b
institution Kabale University
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-8faa06e1e3064f6f989cd8dddee70c0b2025-08-20T03:24:37ZengIEEEIEEE Access2169-35362025-01-0113951229513610.1109/ACCESS.2025.357488111017661Novel View Synthesis of Defocused Blur Scenes Based on Neural Radiance FieldsZhaoji Lin0https://orcid.org/0000-0002-3192-6602Yuxin Zheng1Li Yao2https://orcid.org/0000-0003-2930-8407Department of Computer Science and Engineering, Sanjiang University, Nanjing, ChinaSEU-Monash Joint Graduate School (Suzhou), Southeast University, Suzhou, ChinaSchool of Computer Science and Engineering, Southeast University, Nanjing, ChinaIn recent years, neural radiance fields have been widely used in the field of computer graphics due to their excellent reconstruction quality. However, the shooting process in the wild environment is often affected by various internal and external factors, resulting in blurry images. To address the problem of defocus blur in the real world leading to a decrease in the reconstruction quality of neural radiance fields, this paper proposes a new deblurred radiance field and designs a rigid blur kernel based on the depth features of the image frame to model the rigid transformation of light and the weights of the coarse components of color. For the problem of similar two-dimensional coordinates restricting the model to distinguish scene details in the non-focal plane background, a fine sampling weight using multiscale depth feature fusion is further proposed, and a staged optimization strategy is designed. Experimental results show that compared with the state-of-the-art methods, the method proposed in this paper can better recover scene details and generate high-quality images for defocused blur scenes.https://ieeexplore.ieee.org/document/11017661/Deblurringfeature fusionmultilayer perceptronnovel view synthesisneural radiance field
spellingShingle Zhaoji Lin
Yuxin Zheng
Li Yao
Novel View Synthesis of Defocused Blur Scenes Based on Neural Radiance Fields
IEEE Access
Deblurring
feature fusion
multilayer perceptron
novel view synthesis
neural radiance field
title Novel View Synthesis of Defocused Blur Scenes Based on Neural Radiance Fields
title_full Novel View Synthesis of Defocused Blur Scenes Based on Neural Radiance Fields
title_fullStr Novel View Synthesis of Defocused Blur Scenes Based on Neural Radiance Fields
title_full_unstemmed Novel View Synthesis of Defocused Blur Scenes Based on Neural Radiance Fields
title_short Novel View Synthesis of Defocused Blur Scenes Based on Neural Radiance Fields
title_sort novel view synthesis of defocused blur scenes based on neural radiance fields
topic Deblurring
feature fusion
multilayer perceptron
novel view synthesis
neural radiance field
url https://ieeexplore.ieee.org/document/11017661/
work_keys_str_mv AT zhaojilin novelviewsynthesisofdefocusedblurscenesbasedonneuralradiancefields
AT yuxinzheng novelviewsynthesisofdefocusedblurscenesbasedonneuralradiancefields
AT liyao novelviewsynthesisofdefocusedblurscenesbasedonneuralradiancefields