ViPR: Visual-Odometry-aided Pose Regression for 6DoF Camera Localization

Abstract

Visual Odometry (VO) accumulates a positional drift in long-term robot navigation tasks. Although Convolutional Neural Networks (CNNs) improve VO in various aspects, VO still suffers from moving obstacles, discontinuous observation of features, and poor textures or visual information. While recent approaches estimate a 6DoF pose either directly from (a series of) images or by merging depth maps with optical flow (OF), research that combines absolute pose regression with OF is limited. We propose ViPR, a novel modular architecture for long-term 6DoF VO that leverages temporal information and synergies between absolute pose estimates (from PoseNet-like modules) and relative pose estimates (from FlowNet-based modules) by combining both through recurrent layers. Experiments on known datasets and on our own Industry dataset show that our modular design outperforms state of the art in long-term navigation tasks.

arXiv commentConf. on Computer Vision and Pattern Recognition (CVPR): Joint Workshop on Long-Term Visual Localization, Visual Odometry and Geometric and Learning-based SLAM 2020

References in corpus (13)

• ORB-SLAM2: an Open-Source SLAM System for Monocular, Stereo and RGB-D Cameras
• Past, Present, and Future of Simultaneous Localization And Mapping: Towards the Robust-Perception Age
• DeepVO: Towards End-to-End Visual Odometry with Deep Recurrent Convolutional Neural Networks
• Visual-Inertial Monocular SLAM with Map Reuse
• Bayesian Convolutional Neural Networks with Bernoulli Approximate Variational Inference
• maplab: An Open Framework for Research in Visual-inertial Mapping and Localization
• VLocNet++: Deep Multitask Learning for Semantic Visual Localization and Odometry
• InteriorNet: Mega-scale Multi-sensor Photo-realistic Indoor Scenes Dataset
• LS-VO: Learning Dense Optical Subspace for Robust Visual Odometry Estimation
• Keyframe-Based Visual-Inertial Online SLAM with Relocalization
• Benchmarking 6DOF Outdoor Visual Localization in Changing Conditions
• Improved Visual Relocalization by Discovering Anchor Points
• Image-based Localization using Hourglass Networks

Cited by in corpus (6)

• Fusing Structure from Motion and Simulation-Augmented Pose Regression from Optical Flow for Challenging Indoor Environments
• PixSelect: Less but Reliable Pixels for Accurate and Efficient Localization
• Implicit Learning of Scene Geometry from Poses for Global Localization
• Global Localization: Utilizing Relative Spatio-Temporal Geometric Constraints from Adjacent and Distant Cameras
• PDRNN: Modular Data-driven Pedestrian Dead Reckoning on Loosely Coupled Radio- and Inertial-Signalstreams
• Exploitation of Hidden Context in Dynamic Movement Forecasting: A Neural Network Journey from Recurrent to Graph Neural Networks and General Purpose Transformers