Markov Localisation using Heatmap Regression and Deep Convolutional Odometry
Abstract:
In the context of self-driving vehicles there is strong competition between approaches based on visual localisation and Light Detection And Ranging (LiDAR). While LiDAR provides important depth information, it is sparse in resolution and expensive. On the other hand, cameras are low-cost and recent developments in deep learning mean they can provide high localisation performance. However, several fundamental problems remain, particularly in the domain of uncertainty, where learning based approaches can be notoriously over-confident. Markov, or grid-based, localisation was an early solution to the localisation problem but fell out of favour due to its computational complexity. Representing the likelihood field as a grid (or volume) means there is a trade off between accuracy and memory size. Furthermore, it is necessary to perform expensive convolutions across the entire likelihood volume. Despite the benefit of simultaneously maintaining a likelihood for all possible locations, grid based approaches were superseded by more efficient particle filters and Monte Carlo sampling (MCL). However, MCL introduces its own problems e.g. particle deprivation. Recent advances in deep learning hardware allow large likelihood volumes to be stored directly on the GPU, along with the hardware necessary to efficiently perform GPU-bound 3D convolutions and this obviates many of the disadvantages of grid based methods. In this work, we present a novel CNN-based localisation approach that can leverage modern deep learning hardware. By implementing a grid-based Markov localisation approach directly on the GPU, we create a hybrid Convolutional Neural Network (CNN) that can perform image-based localisation and odometry-based likelihood propagation within a single neural network. The resulting approach is capable of outperforming direct pose regression methods as well as state-of-the-art localisation systems.
Markov Localisation using Heatmap Regression and Deep Convolutional Odometry
Oscar Mendez, Simon Hadfield, Richard Bowden
IEEE International Conference on Robotics and Automation (ICRA 2021)
Project: https://avp-project.uk/
Paper: https://cvssp.org/Personal/OscarMendez/papers/pdf/MendezICRA2021.pdf
Видео Markov Localisation using Heatmap Regression and Deep Convolutional Odometry канала CVSSP Research
In the context of self-driving vehicles there is strong competition between approaches based on visual localisation and Light Detection And Ranging (LiDAR). While LiDAR provides important depth information, it is sparse in resolution and expensive. On the other hand, cameras are low-cost and recent developments in deep learning mean they can provide high localisation performance. However, several fundamental problems remain, particularly in the domain of uncertainty, where learning based approaches can be notoriously over-confident. Markov, or grid-based, localisation was an early solution to the localisation problem but fell out of favour due to its computational complexity. Representing the likelihood field as a grid (or volume) means there is a trade off between accuracy and memory size. Furthermore, it is necessary to perform expensive convolutions across the entire likelihood volume. Despite the benefit of simultaneously maintaining a likelihood for all possible locations, grid based approaches were superseded by more efficient particle filters and Monte Carlo sampling (MCL). However, MCL introduces its own problems e.g. particle deprivation. Recent advances in deep learning hardware allow large likelihood volumes to be stored directly on the GPU, along with the hardware necessary to efficiently perform GPU-bound 3D convolutions and this obviates many of the disadvantages of grid based methods. In this work, we present a novel CNN-based localisation approach that can leverage modern deep learning hardware. By implementing a grid-based Markov localisation approach directly on the GPU, we create a hybrid Convolutional Neural Network (CNN) that can perform image-based localisation and odometry-based likelihood propagation within a single neural network. The resulting approach is capable of outperforming direct pose regression methods as well as state-of-the-art localisation systems.
Markov Localisation using Heatmap Regression and Deep Convolutional Odometry
Oscar Mendez, Simon Hadfield, Richard Bowden
IEEE International Conference on Robotics and Automation (ICRA 2021)
Project: https://avp-project.uk/
Paper: https://cvssp.org/Personal/OscarMendez/papers/pdf/MendezICRA2021.pdf
Видео Markov Localisation using Heatmap Regression and Deep Convolutional Odometry канала CVSSP Research
Показать
Комментарии отсутствуют
Информация о видео
Другие видео канала
3.4 AI@Surrey Summer School - Applications of AI. Medical image databases by Dr Mark Halling-BrownEVReflex: Dense Time-to-Impact Prediction for Event-based Obstacle AvoidanceThe Future of Robotics and AI - CVSSP Robot Lab & Surrey Institute for People Centred AIDigital mental health in adults and older people - Professor Payam BarnaghiS3A SB 3.1. An introduction to the reverb objectS3A SB 3.3. Object-based parameterisation of reverb03. Jaime Spencer - The Monocular Depth Estimation Challenge | 1st MDEC @ WACV23Exploiting high level scene cues in stereo reconstructionBACKUP/ 1.1 AI@Surrey Summer School - Keynote introduction by Prof. Jim Al-Khalili2.1 AI@Surrey Summer School - Basics of deep learning by Prof. John CollomosseDeep4D: A Compact Generative Representation for Volumetric Video05. Chaoqiang Zhao - MonoViT | 1st MDEC @ WACV23SeDAR – Semantic Detection and Ranging: Humans can localise without LiDAR, can robots? ICRA'18Deep 4D Shape Representation: Learning 4D Volumetric Video from Skeletal MotionS3A SB 4.4. Generalised framework for multi-loudspeaker amplitude panning1.1 AI@Surrey Summer School - Keynote introduction by Prof. Jim Al-Khalili4D Temporally Coherent Light-field VideoDECaDE Webinar / Tutorial - Distributed Ledger Technology: Beyond BlockchainS3A SB 2.1. Diffuseness and decorrelationCVMP 2022: High-Quality Variable Speed Video for Stationary Cycling