Neural Radiance Field Dynamic Scene SLAM Based on Ray Segmentation and Bundle Adjustment

The current neural implicit SLAM methods have demonstrated excellent performance in reconstructing ideal static 3D scenes. However, it remains a significant challenge for these methods to handle real scenes with drastic changes in lighting conditions and dynamic environments. This paper proposes a n...

Full description

Saved in:
Bibliographic Details
Main Authors: Yuquan Zhang, Guosheng Feng
Format: Article
Language:English
Published: MDPI AG 2025-03-01
Series:Sensors
Subjects:
Online Access:https://www.mdpi.com/1424-8220/25/6/1679
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850280394440572928
author Yuquan Zhang
Guosheng Feng
author_facet Yuquan Zhang
Guosheng Feng
author_sort Yuquan Zhang
collection DOAJ
description The current neural implicit SLAM methods have demonstrated excellent performance in reconstructing ideal static 3D scenes. However, it remains a significant challenge for these methods to handle real scenes with drastic changes in lighting conditions and dynamic environments. This paper proposes a neural implicit SLAM method that effectively deals with dynamic scenes. We employ a keyframe selection and tracking switching approach based on Lucas–Kanade (LK) optical flow, which serves as prior construction for the Conditional Random Fields potential function. This forms a semantic-based joint estimation method for dynamic and static pixels and constructs corresponding loss functions to impose constraints on dynamic scenes. We conduct experiments on various dynamic and challenging scene datasets, including TUM RGB-D, Openloris, and Bonn. The results demonstrate that our method significantly outperforms existing neural implicit SLAM systems in terms of reconstruction quality and tracking accuracy.
format Article
id doaj-art-0dcaed798afa4b4483959d3b9b89c607
institution OA Journals
issn 1424-8220
language English
publishDate 2025-03-01
publisher MDPI AG
record_format Article
series Sensors
spelling doaj-art-0dcaed798afa4b4483959d3b9b89c6072025-08-20T01:48:46ZengMDPI AGSensors1424-82202025-03-01256167910.3390/s25061679Neural Radiance Field Dynamic Scene SLAM Based on Ray Segmentation and Bundle AdjustmentYuquan Zhang0Guosheng Feng1School of Traffic and Transportation, Shijiazhuang Tiedao University, Shijiazhuang 050043, ChinaSchool of Traffic and Transportation, Shijiazhuang Tiedao University, Shijiazhuang 050043, ChinaThe current neural implicit SLAM methods have demonstrated excellent performance in reconstructing ideal static 3D scenes. However, it remains a significant challenge for these methods to handle real scenes with drastic changes in lighting conditions and dynamic environments. This paper proposes a neural implicit SLAM method that effectively deals with dynamic scenes. We employ a keyframe selection and tracking switching approach based on Lucas–Kanade (LK) optical flow, which serves as prior construction for the Conditional Random Fields potential function. This forms a semantic-based joint estimation method for dynamic and static pixels and constructs corresponding loss functions to impose constraints on dynamic scenes. We conduct experiments on various dynamic and challenging scene datasets, including TUM RGB-D, Openloris, and Bonn. The results demonstrate that our method significantly outperforms existing neural implicit SLAM systems in terms of reconstruction quality and tracking accuracy.https://www.mdpi.com/1424-8220/25/6/1679Dense SLAMneural implicit codingsurface rendering
spellingShingle Yuquan Zhang
Guosheng Feng
Neural Radiance Field Dynamic Scene SLAM Based on Ray Segmentation and Bundle Adjustment
Sensors
Dense SLAM
neural implicit coding
surface rendering
title Neural Radiance Field Dynamic Scene SLAM Based on Ray Segmentation and Bundle Adjustment
title_full Neural Radiance Field Dynamic Scene SLAM Based on Ray Segmentation and Bundle Adjustment
title_fullStr Neural Radiance Field Dynamic Scene SLAM Based on Ray Segmentation and Bundle Adjustment
title_full_unstemmed Neural Radiance Field Dynamic Scene SLAM Based on Ray Segmentation and Bundle Adjustment
title_short Neural Radiance Field Dynamic Scene SLAM Based on Ray Segmentation and Bundle Adjustment
title_sort neural radiance field dynamic scene slam based on ray segmentation and bundle adjustment
topic Dense SLAM
neural implicit coding
surface rendering
url https://www.mdpi.com/1424-8220/25/6/1679
work_keys_str_mv AT yuquanzhang neuralradiancefielddynamicsceneslambasedonraysegmentationandbundleadjustment
AT guoshengfeng neuralradiancefielddynamicsceneslambasedonraysegmentationandbundleadjustment