computer-vision-5

Monocular Visual Odometry

3260 papers • 126 benchmarks • 313 datasets

This task has no description! Would you like to contribute one?

(Image credit: Papersgraph)

Benchmarks

These leaderboards are used to track progress in monocular-visual-odometry-10

Trend

Dataset

Best Model

Actions

No benchmarks available.

Libraries

i

Use these libraries to find monocular-visual-odometry-10 models and implementations

Datasets

Bike and Car Odometer Dataset ! Speedometer OCR

Subtasks

No subtasks available.

Most implemented papers

DeepVO: Towards end-to-end visual odometry with deep Recurrent Convolutional Neural Networks

R. Clark, A. Trigoni, Sen Wang, Hongkai Wen•Sat Jan 14 2017

Extensive experiments on the KITTI VO dataset show competitive performance to state-of-the-art methods, verifying that the end-to-end Deep Learning technique can be a viable complement to the traditional VO systems.

864

Content

0

Paper Graph

Visual Odometry Revisited: What Should Be Learnt?

I. Reid, Jiawang Bian, Huangying Zhan, C. Weerasekera•Fri Sep 20 2019

This work revisit the basics of VO and explore the right way for integrating deep learning with epipolar geometry and Perspective-n-Point method and design a simple but robust frame-to-frame VO algorithm (DF-VO) which outperforms pure deep learning-based and geometry-based methods.

184 0

Paper Graph

DF-VO: What Should Be Learnt for Visual Odometry?

I. Reid, Jiawang Bian, Huangying Zhan, C. Weerasekera, Ravi Garg•Sun Feb 28 2021

This work proposes a method to carefully sample high-quality correspondences from deep flows and recover accurate camera poses with a geometric module, and addresses the scale-drift issue by aligning geometrically triangulated depths to thescale-consistent deep depths, where the dynamic scenes are taken into account.

54 0

Paper Graph

Unsupervised Scale-Consistent Depth Learning from Video

Naiyan Wang, I. Reid, Chunhua Shen, Ming-Ming Cheng, Zhichao Li, Le Zhang, Jiawang Bian, Huangying Zhan•Mon May 24 2021

A monocular depth estimation method SC-Depth, which requires only unlabelled videos for training and enables the scale-consistent prediction at inference time and the proposed hybrid Pseudo-RGBD SLAM shows compelling results in KITTI, and it generalizes well to the KAIST dataset without additional training.

208 0

Paper Graph

Sparse Representations for Object and Ego-motion Estimation in Dynamic Scenes

J. Krichmar, H. Kashyap, C. Fowlkes•Fri Mar 08 2019

A learning based approach to predict camera motion parameters directly from optic flow, by marginalizing depthmap variations and outliers by learning a sparse overcomplete basis set of egomotion in an autoencoder network.

3 0

Paper Graph

Extending Monocular Visual Odometry to Stereo Camera Systems by Scale optimization

Junaed Sattar, Jiawei Mo•Tue May 28 2019

The proposed method uses an additional camera to accurately estimate and optimize the scale of the monocular visual odometry, rather than triangulating 3D points from stereo matching, and it is computationally efficient, adding minimal overhead to the stereo vision system compared to straightforward stereo matching.

11 0

Paper Graph

EndoSLAM Dataset and An Unsupervised Monocular Visual Odometry and Depth Estimation Approach for Endoscopic Videos: Endo-SfMLearner

K. Ozyoruk, Guliz Irem Gokceler, Gulfize Coskun, Kagan Incetan, Yasin Almalioglu, Faisal Mahmood, E. Curto, Luis Perdigoto, Marina Oliveira, Hasan Sahin, Helder Araújo, Henrique Alexandrino, Nicholas J. Durr, H. Gilbert, Mehmet Turan•Mon Jun 29 2020

A comprehensive endoscopic SLAM dataset consisting of 3D point cloud data for six porcine organs, capsule and standard endoscopy recordings as well as synthetically generated data is introduced and Endo-SfMLearner, an unsupervised monocular depth and pose estimation method that combines residual networks with spatial attention module is propound.

15 0

Paper Graph

WGANVO: Monocular Visual Odometry based on Generative Adversarial Networks

Javier Cremona, Lucas C. Uzal, Taihú Pire•Sun Jul 26 2020

This work presents WGANVO, a Deep Learning based monocular Visual Odometry method, where a neural network is trained to regress a pose estimate from an image pair using a semi-supervised approach.

7 0

Paper Graph

Instant Visual Odometry Initialization for Mobile AR

Jesus Briales, Christian Forster, Alejo Concha, M. Burri, Luc Oth•Thu Jul 29 2021

This paper presents a 6-DoF monocular visual odometry that initializes instantly and without motion parallax, and shows that the proposed pose estimator outperforms the classical approaches for 6- doF pose estimation used in the literature in low-parallax configurations.

14 0

Paper Graph

Adding a benchmark result helps the community track progress.