Pixel-Wise Prediction based Visual Odometry via Uncertainty Estimation
Not on the map. Why?
Abstract (arXiv)
This paper introduces pixel-wise prediction based visual odometry (PWVO), which is a dense prediction task that evaluates the values of translation and rotation for every pixel in its input observations. PWVO employs uncertainty estimation to identify the noisy regions in the input observations, and adopts a selection mechanism to integrate pixel-wise predictions based on the estimated uncertainty maps to derive the final translation and rotation. In order to train PWVO in a comprehensive fashion, we further develop a data generation workflow for generating synthetic training data. The experimental results show that PWVO is able to deliver favorable results. In addition, our analyses validate the effectiveness of the designs adopted in PWVO, and demonstrate that the uncertainty maps estimated by PWVO is capable of capturing the noises in its input observations.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 32 of its 60 references that are in the map. First posted on arXiv 18 Aug 2022.
Builds on
32 of 60 references.
References · 32
- 2013Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Mapscited by 2,348
- 2017What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision?cited by 2,335
- 2016ORB-SLAM2: an Open-Source SLAM System for Monocular, Stereo and RGB-D Camerascited by 801
- 2017PoseCNN: A Convolutional Neural Network for 6D Object Pose Estimation in Cluttered Scenescited by 768
- 2020TartanAir: A Dataset to Push the Limits of Visual SLAMcited by 494
- 2016Direct Sparse Odometrycited by 479
- 2018An Intriguing Failing of Convolutional Neural Networks and the CoordConv Solutioncited by 302
- 2017Geometric Loss Functions for Camera Pose Regression with Deep Learningcited by 224
- 2020D3VO: Deep Depth, Deep Pose and Deep Uncertainty for Monocular Visual Odometrycited by 206
- 2019Understanding the Limitations of CNN-based Absolute Camera Pose Regressioncited by 177
- 2017DeepVO: Towards End-to-End Visual Odometry with Deep Recurrent Convolutional Neural Networkscited by 173
- 2017Geometry-Aware Learning of Maps for Camera Localizationcited by 161
20 more
- 2016Image-based localization using LSTMs for structured feature correlationcited by 136
- 2015Modelling Uncertainty in Deep Learning for Camera Relocalizationcited by 133
- 2018DynaSLAM: Tracking, Mapping and Inpainting in Dynamic Scenescited by 130
- 2020TartanVO: A Generalizable Learning-based VOcited by 107
- 2015Learning to See by Movingcited by 101
- 2019AtLoc: Attention Guided Camera Localizationcited by 92
- 2017Camera Relocalization by Computing Pairwise Relative Poses Using Convolutional Neural Networkcited by 79
- 2017Image-based Localization using Hourglass Networkscited by 63
- 2018VLocNet++: Deep Multitask Learning for Semantic Visual Localization and Odometrycited by 61
- 2018Deep Auxiliary Learning for Visual Localization and Odometrycited by 59
- 2017Relative Camera Pose Estimation Using Convolutional Neural Networkscited by 47
- 2015Learning image representations tied to ego-motioncited by 41
- 2019Selective Sensor Fusion for Neural Visual-Inertial Odometrycited by 34
- 2021Attentive and Contrastive Learning for Joint Depth and Motion Field Estimationcited by 24
- 2020Attentional Separation-and-Aggregation Network for Self-supervised Depth-Pose Learning in Dynamic Scenescited by 11
- 2018Global Pose Estimation with an Attention-based Recurrent Networkcited by 10
- 2018Guided Feature Selection for Deep Visual Odometrycited by 10
- 2017LS-VO: Learning Dense Optical Subspace for Robust Visual Odometry Estimationcited by 7
- 2020Exploring Self-Attention for Visual Odometrycited by 3
- 2019Movable-Object-Aware Visual SLAM via Weakly Supervised Semantic Segmentationcited by 2
Built on by
Citing papers per year: 2024: 1.
1 paper.