Get our free extension to see links to code for papers anywhere online!

Chrome logo Add to Chrome

Firefox logo Add to Firefox
Picture for Michael Milford

Michael Milford

Zero-Shot Domain Adaptation with a Physics Prior


Aug 11, 2021
Attila Lengyel, Sourav Garg, Michael Milford, Jan C. van Gemert

Figure 1 for Zero-Shot Domain Adaptation with a Physics Prior
Figure 2 for Zero-Shot Domain Adaptation with a Physics Prior
Figure 3 for Zero-Shot Domain Adaptation with a Physics Prior
Figure 4 for Zero-Shot Domain Adaptation with a Physics Prior
* ICCV 2021 Oral presentation. Code, datasets and supplementary material: https://github.com/Attila94/CIConv 
  Access Paper or Ask Questions

Bayesian Controller Fusion: Leveraging Control Priors in Deep Reinforcement Learning for Robotics


Jul 22, 2021
Krishan Rana, Vibhavari Dasagi, Jesse Haviland, Ben Talbot, Michael Milford, Niko Sünderhauf

Figure 1 for Bayesian Controller Fusion: Leveraging Control Priors in Deep Reinforcement Learning for Robotics
Figure 2 for Bayesian Controller Fusion: Leveraging Control Priors in Deep Reinforcement Learning for Robotics
Figure 3 for Bayesian Controller Fusion: Leveraging Control Priors in Deep Reinforcement Learning for Robotics
Figure 4 for Bayesian Controller Fusion: Leveraging Control Priors in Deep Reinforcement Learning for Robotics
* Under review for The International Journal of Robotics Research (IJRR). Project page: https://krishanrana.github.io/bcf 
  Access Paper or Ask Questions

Probabilistic Appearance-Invariant Topometric Localization with New Place Awareness


Jul 16, 2021
Ming Xu, Tobias Fischer, Niko Sünderhauf, Michael Milford

Figure 1 for Probabilistic Appearance-Invariant Topometric Localization with New Place Awareness
Figure 2 for Probabilistic Appearance-Invariant Topometric Localization with New Place Awareness
Figure 3 for Probabilistic Appearance-Invariant Topometric Localization with New Place Awareness
Figure 4 for Probabilistic Appearance-Invariant Topometric Localization with New Place Awareness
* IEEE Robotics and Automation Letters and IROS 2021 
 * 8 pages 
  Access Paper or Ask Questions

A Hierarchical Dual Model of Environment- and Place-Specific Utility for Visual Place Recognition


Jul 06, 2021
Nikhil Varma Keetha, Michael Milford, Sourav Garg

Figure 1 for A Hierarchical Dual Model of Environment- and Place-Specific Utility for Visual Place Recognition
Figure 2 for A Hierarchical Dual Model of Environment- and Place-Specific Utility for Visual Place Recognition
Figure 3 for A Hierarchical Dual Model of Environment- and Place-Specific Utility for Visual Place Recognition
Figure 4 for A Hierarchical Dual Model of Environment- and Place-Specific Utility for Visual Place Recognition
* Accepted to IEEE Robotics and Automation Letters (RA-L) and IROS 2021 
  Access Paper or Ask Questions

SeqNetVLAD vs PointNetVLAD: Image Sequence vs 3D Point Clouds for Day-Night Place Recognition


Jun 22, 2021
Sourav Garg, Michael Milford

Figure 1 for SeqNetVLAD vs PointNetVLAD: Image Sequence vs 3D Point Clouds for Day-Night Place Recognition
Figure 2 for SeqNetVLAD vs PointNetVLAD: Image Sequence vs 3D Point Clouds for Day-Night Place Recognition
Figure 3 for SeqNetVLAD vs PointNetVLAD: Image Sequence vs 3D Point Clouds for Day-Night Place Recognition
* Accepted to CVPR 2021 Workshop on 3D Vision and Robotics (3DVR). https://sites.google.com/view/cvpr2021-3d-vision-robotics/ 
  Access Paper or Ask Questions

Probabilistic Visual Place Recognition for Hierarchical Localization


May 07, 2021
Ming Xu, Niko Sünderhauf, Michael Milford

Figure 1 for Probabilistic Visual Place Recognition for Hierarchical Localization
Figure 2 for Probabilistic Visual Place Recognition for Hierarchical Localization
Figure 3 for Probabilistic Visual Place Recognition for Hierarchical Localization
Figure 4 for Probabilistic Visual Place Recognition for Hierarchical Localization
* 8 pages, 4 figures, RA-L standalone 
  Access Paper or Ask Questions

RoboStack: Using the Robot Operating System alongside the Conda and Jupyter Data Science Ecosystems


Apr 26, 2021
Tobias Fischer, Wolf Vollprecht, Silvio Traversaro, Sean Yen, Carlos Herrero, Michael Milford

Figure 1 for RoboStack: Using the Robot Operating System alongside the Conda and Jupyter Data Science Ecosystems
Figure 2 for RoboStack: Using the Robot Operating System alongside the Conda and Jupyter Data Science Ecosystems
Figure 3 for RoboStack: Using the Robot Operating System alongside the Conda and Jupyter Data Science Ecosystems
Figure 4 for RoboStack: Using the Robot Operating System alongside the Conda and Jupyter Data Science Ecosystems
* Under review 
  Access Paper or Ask Questions

Uncertainty for Identifying Open-Set Errors in Visual Object Detection


Apr 03, 2021
Dimity Miller, Niko Sünderhauf, Michael Milford, Feras Dayoub

Figure 1 for Uncertainty for Identifying Open-Set Errors in Visual Object Detection
Figure 2 for Uncertainty for Identifying Open-Set Errors in Visual Object Detection
Figure 3 for Uncertainty for Identifying Open-Set Errors in Visual Object Detection
Figure 4 for Uncertainty for Identifying Open-Set Errors in Visual Object Detection

  Access Paper or Ask Questions

RoRD: Rotation-Robust Descriptors and Orthographic Views for Local Feature Matching


Mar 15, 2021
Udit Singh Parihar, Aniket Gujarathi, Kinal Mehta, Satyajit Tourani, Sourav Garg, Michael Milford, K. Madhava Krishna

Figure 1 for RoRD: Rotation-Robust Descriptors and Orthographic Views for Local Feature Matching
Figure 2 for RoRD: Rotation-Robust Descriptors and Orthographic Views for Local Feature Matching
Figure 3 for RoRD: Rotation-Robust Descriptors and Orthographic Views for Local Feature Matching
Figure 4 for RoRD: Rotation-Robust Descriptors and Orthographic Views for Local Feature Matching
* Under Review at IROS. Project Page: https://uditsinghparihar.github.io/RoRD/ 
  Access Paper or Ask Questions

Where is your place, Visual Place Recognition?


Mar 11, 2021
Sourav Garg, Tobias Fischer, Michael Milford

Figure 1 for Where is your place, Visual Place Recognition?
* Under Review 
  Access Paper or Ask Questions

Sequential Place Learning: Heuristic-Free High-Performance Long-Term Place Recognition


Mar 02, 2021
Marvin Chancán, Michael Milford

Figure 1 for Sequential Place Learning: Heuristic-Free High-Performance Long-Term Place Recognition
Figure 2 for Sequential Place Learning: Heuristic-Free High-Performance Long-Term Place Recognition
Figure 3 for Sequential Place Learning: Heuristic-Free High-Performance Long-Term Place Recognition
Figure 4 for Sequential Place Learning: Heuristic-Free High-Performance Long-Term Place Recognition
* Submitted to RSS 2021. 14 pages, 5 tables, 17 figures 
  Access Paper or Ask Questions

Sequence-Based Filtering for Visual Route-Based Navigation: Analysing the Benefits, Trade-offs and Design Choices


Mar 02, 2021
Mihnea-Alexandru Tomită, Mubariz Zaffar, Michael Milford, Klaus McDonald-Maier, Shoaib Ehsan

Figure 1 for Sequence-Based Filtering for Visual Route-Based Navigation: Analysing the Benefits, Trade-offs and Design Choices
Figure 2 for Sequence-Based Filtering for Visual Route-Based Navigation: Analysing the Benefits, Trade-offs and Design Choices
Figure 3 for Sequence-Based Filtering for Visual Route-Based Navigation: Analysing the Benefits, Trade-offs and Design Choices
Figure 4 for Sequence-Based Filtering for Visual Route-Based Navigation: Analysing the Benefits, Trade-offs and Design Choices
* 7 pages, currently under-review 
  Access Paper or Ask Questions

Patch-NetVLAD: Multi-Scale Fusion of Locally-Global Descriptors for Place Recognition


Mar 02, 2021
Stephen Hausler, Sourav Garg, Ming Xu, Michael Milford, Tobias Fischer

Figure 1 for Patch-NetVLAD: Multi-Scale Fusion of Locally-Global Descriptors for Place Recognition
Figure 2 for Patch-NetVLAD: Multi-Scale Fusion of Locally-Global Descriptors for Place Recognition
Figure 3 for Patch-NetVLAD: Multi-Scale Fusion of Locally-Global Descriptors for Place Recognition
Figure 4 for Patch-NetVLAD: Multi-Scale Fusion of Locally-Global Descriptors for Place Recognition
* Accepted to IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2021) 
  Access Paper or Ask Questions

Scene Retrieval for Contextual Visual Mapping


Feb 25, 2021
William H. B. Smith, Michael Milford, Klaus D. McDonald-Maier, Shoaib Ehsan

Figure 1 for Scene Retrieval for Contextual Visual Mapping
Figure 2 for Scene Retrieval for Contextual Visual Mapping
Figure 3 for Scene Retrieval for Contextual Visual Mapping
Figure 4 for Scene Retrieval for Contextual Visual Mapping
* 8 page paper on visual place recogniton and scene classification 
  Access Paper or Ask Questions

SeqNet: Learning Descriptors for Sequence-based Hierarchical Place Recognition


Feb 24, 2021
Sourav Garg, Michael Milford

Figure 1 for SeqNet: Learning Descriptors for Sequence-based Hierarchical Place Recognition
Figure 2 for SeqNet: Learning Descriptors for Sequence-based Hierarchical Place Recognition
Figure 3 for SeqNet: Learning Descriptors for Sequence-based Hierarchical Place Recognition
Figure 4 for SeqNet: Learning Descriptors for Sequence-based Hierarchical Place Recognition
* Accepted for publication in IEEE RA-L 2021; includes supplementary 
  Access Paper or Ask Questions

Improving Visual Place Recognition Performance by Maximising Complementarity


Feb 16, 2021
Maria Waheed, Michael Milford, Klaus D. McDonald-Maier, Shoaib Ehsan

Figure 1 for Improving Visual Place Recognition Performance by Maximising Complementarity
Figure 2 for Improving Visual Place Recognition Performance by Maximising Complementarity
Figure 3 for Improving Visual Place Recognition Performance by Maximising Complementarity
Figure 4 for Improving Visual Place Recognition Performance by Maximising Complementarity

  Access Paper or Ask Questions

Semantics for Robotic Mapping, Perception and Interaction: A Survey


Jan 02, 2021
Sourav Garg, Niko Sünderhauf, Feras Dayoub, Douglas Morrison, Akansel Cosgun, Gustavo Carneiro, Qi Wu, Tat-Jun Chin, Ian Reid, Stephen Gould, Peter Corke, Michael Milford

Figure 1 for Semantics for Robotic Mapping, Perception and Interaction: A Survey
Figure 2 for Semantics for Robotic Mapping, Perception and Interaction: A Survey
* Foundations and Trends in Robotics: Vol. 8: No. 1-2, pp 1-224 (2020) 
 * 81 pages, 1 figure, published in Foundations and Trends in Robotics, 2020 
  Access Paper or Ask Questions

DeepSeqSLAM: A Trainable CNN+RNN for Joint Global Description and Sequence-based Place Recognition


Nov 17, 2020
Marvin Chancán, Michael Milford

Figure 1 for DeepSeqSLAM: A Trainable CNN+RNN for Joint Global Description and Sequence-based Place Recognition
Figure 2 for DeepSeqSLAM: A Trainable CNN+RNN for Joint Global Description and Sequence-based Place Recognition
Figure 3 for DeepSeqSLAM: A Trainable CNN+RNN for Joint Global Description and Sequence-based Place Recognition
Figure 4 for DeepSeqSLAM: A Trainable CNN+RNN for Joint Global Description and Sequence-based Place Recognition
* NeurIPS 2020 Workshop on Machine Learning for Autonomous Driving (ML4AD) 
 * 9 pages, 6 figures, 2 tables 
  Access Paper or Ask Questions

Fast and Robust Bio-inspired Teach and Repeat Navigation


Oct 21, 2020
Dominic Dall'Osto, Tobias Fischer, Michael Milford

Figure 1 for Fast and Robust Bio-inspired Teach and Repeat Navigation
Figure 2 for Fast and Robust Bio-inspired Teach and Repeat Navigation
Figure 3 for Fast and Robust Bio-inspired Teach and Repeat Navigation
Figure 4 for Fast and Robust Bio-inspired Teach and Repeat Navigation
* 8 pages, 8 figures, paper is currently under review 
  Access Paper or Ask Questions

Intelligent Reference Curation for Visual Place Recognition via Bayesian Selective Fusion


Oct 19, 2020
Timothy L. Molloy, Tobias Fischer, Michael Milford, Girish N. Nair

Figure 1 for Intelligent Reference Curation for Visual Place Recognition via Bayesian Selective Fusion
Figure 2 for Intelligent Reference Curation for Visual Place Recognition via Bayesian Selective Fusion
Figure 3 for Intelligent Reference Curation for Visual Place Recognition via Bayesian Selective Fusion
Figure 4 for Intelligent Reference Curation for Visual Place Recognition via Bayesian Selective Fusion
* 8 pages, 10 figures, paper is currently under review 
  Access Paper or Ask Questions

Binary Neural Networks for Memory-Efficient and Effective Visual Place Recognition in Changing Environments


Oct 01, 2020
Bruno Ferrarini, Michael Milford, Klaus D. McDonald-Maier, Shoaib Ehsan

Figure 1 for Binary Neural Networks for Memory-Efficient and Effective Visual Place Recognition in Changing Environments
Figure 2 for Binary Neural Networks for Memory-Efficient and Effective Visual Place Recognition in Changing Environments
Figure 3 for Binary Neural Networks for Memory-Efficient and Effective Visual Place Recognition in Changing Environments
Figure 4 for Binary Neural Networks for Memory-Efficient and Effective Visual Place Recognition in Changing Environments

  Access Paper or Ask Questions

ConvSequential-SLAM: A Sequence-based, Training-less Visual Place Recognition Technique for Changing Environments


Sep 28, 2020
Mihnea-Alexandru Tomită, Mubariz Zaffar, Michael Milford, Klaus McDonald-Maier, Shoaib Ehsan

Figure 1 for ConvSequential-SLAM: A Sequence-based, Training-less Visual Place Recognition Technique for Changing Environments
Figure 2 for ConvSequential-SLAM: A Sequence-based, Training-less Visual Place Recognition Technique for Changing Environments
Figure 3 for ConvSequential-SLAM: A Sequence-based, Training-less Visual Place Recognition Technique for Changing Environments
Figure 4 for ConvSequential-SLAM: A Sequence-based, Training-less Visual Place Recognition Technique for Changing Environments
* 10 pages, currently under-review 
  Access Paper or Ask Questions

Robot Perception enables Complex Navigation Behavior via Self-Supervised Learning


Jun 16, 2020
Marvin Chancán, Michael Milford

Figure 1 for Robot Perception enables Complex Navigation Behavior via Self-Supervised Learning
Figure 2 for Robot Perception enables Complex Navigation Behavior via Self-Supervised Learning
Figure 3 for Robot Perception enables Complex Navigation Behavior via Self-Supervised Learning
Figure 4 for Robot Perception enables Complex Navigation Behavior via Self-Supervised Learning
* Submitted to RSS 2020 Workshop on Self-Supervised Robot Learning 
  Access Paper or Ask Questions

Delta Descriptors: Change-Based Place Representation for Robust Visual Localization


Jun 10, 2020
Sourav Garg, Ben Harwood, Gaurangi Anand, Michael Milford

Figure 1 for Delta Descriptors: Change-Based Place Representation for Robust Visual Localization
Figure 2 for Delta Descriptors: Change-Based Place Representation for Robust Visual Localization
Figure 3 for Delta Descriptors: Change-Based Place Representation for Robust Visual Localization
Figure 4 for Delta Descriptors: Change-Based Place Representation for Robust Visual Localization
* 8 pages and 7 figures. To be published in 2020 IEEE Robotics and Automation Letters (RA-L) 
  Access Paper or Ask Questions

Event-based visual place recognition with ensembles of spatio-temporal windows


May 22, 2020
Tobias Fischer, Michael Milford

Figure 1 for Event-based visual place recognition with ensembles of spatio-temporal windows
Figure 2 for Event-based visual place recognition with ensembles of spatio-temporal windows
Figure 3 for Event-based visual place recognition with ensembles of spatio-temporal windows
Figure 4 for Event-based visual place recognition with ensembles of spatio-temporal windows
* 8 pages, 8 figures, paper is currently under review 
  Access Paper or Ask Questions

VPR-Bench: An Open-Source Visual Place Recognition Evaluation Framework with Quantifiable Viewpoint and Appearance Change


May 17, 2020
Mubariz Zaffar, Shoaib Ehsan, Michael Milford, David Flynn, Klaus McDonald-Maier

Figure 1 for VPR-Bench: An Open-Source Visual Place Recognition Evaluation Framework with Quantifiable Viewpoint and Appearance Change
Figure 2 for VPR-Bench: An Open-Source Visual Place Recognition Evaluation Framework with Quantifiable Viewpoint and Appearance Change
Figure 3 for VPR-Bench: An Open-Source Visual Place Recognition Evaluation Framework with Quantifiable Viewpoint and Appearance Change
Figure 4 for VPR-Bench: An Open-Source Visual Place Recognition Evaluation Framework with Quantifiable Viewpoint and Appearance Change
* Currently under-review, 25 pages, 16 figures 
  Access Paper or Ask Questions

Class Anchor Clustering: a Distance-based Loss for Training Open Set Classifiers


Apr 06, 2020
Dimity Miller, Niko Sünderhauf, Michael Milford, Feras Dayoub

Figure 1 for Class Anchor Clustering: a Distance-based Loss for Training Open Set Classifiers
Figure 2 for Class Anchor Clustering: a Distance-based Loss for Training Open Set Classifiers
Figure 3 for Class Anchor Clustering: a Distance-based Loss for Training Open Set Classifiers
Figure 4 for Class Anchor Clustering: a Distance-based Loss for Training Open Set Classifiers

  Access Paper or Ask Questions

Multiplicative Controller Fusion: A Hybrid Navigation Strategy For Deployment in Unknown Environments


Mar 13, 2020
Krishan Rana, Vibhavari Dasagi, Ben Talbot, Michael Milford, Niko Sünderhauf

Figure 1 for Multiplicative Controller Fusion: A Hybrid Navigation Strategy For Deployment in Unknown Environments
Figure 2 for Multiplicative Controller Fusion: A Hybrid Navigation Strategy For Deployment in Unknown Environments
Figure 3 for Multiplicative Controller Fusion: A Hybrid Navigation Strategy For Deployment in Unknown Environments
Figure 4 for Multiplicative Controller Fusion: A Hybrid Navigation Strategy For Deployment in Unknown Environments
* Under review for IROS2020. Project site available at https://sites.google.com/view/mcf-nav/home 
  Access Paper or Ask Questions