Abstract:In numerous robotic and autonomous driving tasks, traditional visual SLAM algorithms estimate the camera's position in a scene through sparse feature points and express the map by estimating the depth of sparse point clouds. However, practical applications require SLAM to create dense maps in real time, overcoming the sparsity and occlusion issues of point clouds. Furthermore, it is advantageous for SLAM map to possess an auto-completion capability, where the map can automatically infer and complete the remaining 20% when the camera observes only 80% of an object. Therefore, a more dense and intelligent map representation is needed. In this paper, we propose a Visual–Inertial SLAM with Neural Radiance Fields reconstruction to address the aforementioned challenges. We integrate the traditional rule-based optimization with NeRF. This approach allows for the real-time update of NeRF local functions by rapidly estimating camera motion and sparse feature point depths to reconstruct 3D scenes. To achieve better camera poses and globally consistent map, we address the issue of IMU noise spikes resulting from rapid motion changes, along with handling pose adjustments due to loop closure fusion. Specifically, we employ a form of widening the static noise covariance to refit the dynamic noise covariance. During loop closure fusion, we treat the pose adjustment between pre- and post-loop closure as a spatiotemporal transformation, migrating NeRF parameters from pre- to post- to expedite loop closure adjustments in NeRF mapping. Moreover, we extend this method to scenarios with only grayscale images. By expanding the color channels of grayscale images and conducting linear spatial mapping, we can rapidly reconstruct 3D scenes with only grayscale images. We demonstrate the precision and speed advantages of our method in both RGB and grayscale scenes.

EC-SLAM: Real-time Dense Neural RGB-D SLAM System with Effectively Constrained Global Bundle Adjustment.

EC-SLAM: Effectively Constrained Neural RGB-D SLAM with Sparse TSDF Encoding and Global Bundle Adjustment

Robust Keyframe-based Dense SLAM with an RGB-D Camera.

SLAIM: Robust Dense Neural SLAM for Online Tracking and Mapping

NeSLAM: Neural Implicit Mapping and Self-Supervised Feature Tracking With Depth Completion and Denoising

ESLAM: Efficient Dense SLAM System Based on Hybrid Representation of Signed Distance Fields

Co-SLAM: Joint Coordinate and Sparse Parametric Encodings for Neural Real-Time SLAM

VI-NeRF-SLAM: a real-time visual–inertial SLAM with NeRF mapping

GlORIE-SLAM: Globally Optimized RGB-only Implicit Encoding Point Cloud SLAM

CG-SLAM: Efficient Dense RGB-D SLAM in a Consistent Uncertainty-aware 3D Gaussian Field

RoDyn-SLAM: Robust Dynamic Dense RGB-D SLAM with Neural Radiance Fields

NICER-SLAM: Neural Implicit Scene Encoding for RGB SLAM

Implicit Event-RGBD Neural SLAM

TivNe-SLAM: Dynamic Mapping and Tracking via Time-Varying Neural Radiance Fields

ONeK-SLAM: A Robust Object-level Dense SLAM Based on Joint Neural Radiance Fields and Keypoints

Point-SLAM: Dense Neural Point Cloud-based SLAM

CP-SLAM: Collaborative Neural Point-based SLAM System

SAR-SLAM: Self-Attentive Rendering-based SLAM with Neural Point Cloud Encoding

Structerf-SLAM: Neural implicit representation SLAM for structural environments

NID-SLAM: Neural Implicit Representation-based RGB-D SLAM in dynamic environments