Matthew Gadd
YOU?
Author Swipe
View article: Biodiversity research requires more rotors and wheels on and above ground, as well as below water
Biodiversity research requires more rotors and wheels on and above ground, as well as below water Open
Human activities have caused rapid decline in biodiversity, with accelerating species extinction. Simultaneously, recent advancements in artificial intelligence and autonomous systems offer transformative potential for biodiversity researc…
View article: Watching Grass Grow: Long-term Visual Navigation and Mission Planning for Autonomous Biodiversity Monitoring
Watching Grass Grow: Long-term Visual Navigation and Mission Planning for Autonomous Biodiversity Monitoring Open
We describe a challenging robotics deployment in a complex ecosystem to monitor a rich plant community. The study site is dominated by dynamic grassland vegetation and is thus visually ambiguous and liable to drastic appearance change over…
View article: VDNA-PR: Using General Dataset Representations for Robust Sequential Visual Place Recognition
VDNA-PR: Using General Dataset Representations for Robust Sequential Visual Place Recognition Open
This paper adapts a general dataset representation technique to produce robust Visual Place Recognition (VPR) descriptors, crucial to enable real-world mobile robot localisation. Two parallel lines of work on VPR have shown, on one side, t…
View article: That's My Point: Compact Object-centric LiDAR Pose Estimation for Large-scale Outdoor Localisation
That's My Point: Compact Object-centric LiDAR Pose Estimation for Large-scale Outdoor Localisation Open
This paper is about 3D pose estimation on LiDAR scans with extremely minimal storage requirements to enable scalable mapping and localisation. We achieve this by clustering all points of segmented scans into semantic objects and representi…
View article: OORD: The Oxford Offroad Radar Dataset
OORD: The Oxford Offroad Radar Dataset Open
There is a growing academic interest as well as commercial exploitation of millimetre-wave scanning radar for autonomous vehicle localisation and scene understanding. Although several datasets to support this research area have been releas…
View article: Masked Gamma-SSL: Learning Uncertainty Estimation via Masked Image Modeling
Masked Gamma-SSL: Learning Uncertainty Estimation via Masked Image Modeling Open
This work proposes a semantic segmentation network that produces high-quality uncertainty estimates in a single forward pass. We exploit general representations from foundation models and unlabelled datasets through a Masked Image Modeling…
View article: Mitigating Distributional Shift in Semantic Segmentation via Uncertainty Estimation from Unlabelled Data
Mitigating Distributional Shift in Semantic Segmentation via Uncertainty Estimation from Unlabelled Data Open
Knowing when a trained segmentation model is encountering data that is different to its training data is important. Understanding and mitigating the effects of this play an important part in their application from a performance and assuran…
View article: RAG-Driver: Generalisable Driving Explanations with Retrieval-Augmented In-Context Learning in Multi-Modal Large Language Model
RAG-Driver: Generalisable Driving Explanations with Retrieval-Augmented In-Context Learning in Multi-Modal Large Language Model Open
We need to trust robots that use often opaque AI methods. They need to explain themselves to us, and we need to trust their explanation. In this regard, explainability plays a critical role in trustworthy autonomous decision-making to fost…
View article: Open-RadVLAD: Fast and Robust Radar Place Recognition
Open-RadVLAD: Fast and Robust Radar Place Recognition Open
Radar place recognition often involves encoding a live scan as a vector and matching this vector to a database in order to recognise that the vehicle is in a location that it has visited before. Radar is inherently robust to lighting or we…
View article: Mitigating Distributional Shift in Semantic Segmentation via Uncertainty Estimation From Unlabeled Data
Mitigating Distributional Shift in Semantic Segmentation via Uncertainty Estimation From Unlabeled Data Open
Knowing when a trained segmentation model is encountering data that is different to its training data is important. Understanding and mitigating the effects of this play an important part in their application from a performance and assuran…
View article: Robot-Relay : Building-Wide, Calibration-Less Visual Servoing with Learned Sensor Handover Network
Robot-Relay : Building-Wide, Calibration-Less Visual Servoing with Learned Sensor Handover Network Open
We present a system which grows and manages a network of remote viewpoints during the natural installation cycle for a newly installed camera network or a newly deployed robot fleet. No explicit notion of camera position or orientation is …
View article: What you see is what you get: Experience ranking with deep neural dataset-to-dataset similarity for topological localisation
What you see is what you get: Experience ranking with deep neural dataset-to-dataset similarity for topological localisation Open
Recalling the most relevant visual memories for localisation or understanding a priori the likely outcome of localisation effort against a particular visual memory is useful for efficient and robust visual navigation. Solutions to this pro…
View article: LROC-PANGU-GAN: Closing the Simulation Gap in Learning Crater Segmentation with Planetary Simulators
LROC-PANGU-GAN: Closing the Simulation Gap in Learning Crater Segmentation with Planetary Simulators Open
It is critical for probes landing on foreign planetary bodies to be able to robustly identify and avoid hazards - as, for example, steep cliffs or deep craters can pose significant risks to a probe's landing and operational success. Recent…
View article: SEM-GAT: Explainable Semantic Pose Estimation using Learned Graph Attention
SEM-GAT: Explainable Semantic Pose Estimation using Learned Graph Attention Open
This paper proposes a Graph Neural Network(GNN)-based method for exploiting semantics and local geometry to guide the identification of reliable pointcloud registration candidates. Semantic and morphological features of the environment ser…
View article: Visual Servoing on Wheels: Robust Robot Orientation Estimation in Remote Viewpoint Control
Visual Servoing on Wheels: Robust Robot Orientation Estimation in Remote Viewpoint Control Open
This work proposes a fast deployment pipeline for visually-servoed robots which does not assume anything about either the robot - e.g. sizes, colour or the presence of markers - or the deployment environment. In this, accurate estimation o…
View article: Off the Radar: Uncertainty-Aware Radar Place Recognition with Introspective Querying and Map Maintenance
Off the Radar: Uncertainty-Aware Radar Place Recognition with Introspective Querying and Map Maintenance Open
Localisation with Frequency-Modulated Continuous-Wave (FMCW) radar has gained increasing interest due to its inherent resistance to challenging environments. However, complex artefacts of the radar measurement process require appropriate u…
View article: Visual DNA: Representing and Comparing Images using Distributions of Neuron Activations
Visual DNA: Representing and Comparing Images using Distributions of Neuron Activations Open
Selecting appropriate datasets is critical in modern computer vision. However, no general-purpose tools exist to evaluate the extent to which two datasets differ. For this, we propose representing images - and by extension datasets - using…
View article: BoxGraph: Semantic Place Recognition and Pose Estimation from 3D LiDAR
BoxGraph: Semantic Place Recognition and Pose Estimation from 3D LiDAR Open
This paper is about extremely robust and lightweight localisation using LiDAR point clouds based on instance segmentation and graph matching. We model 3D point clouds as fully-connected graphs of semantically identified components where ea…
View article: What Goes Around: Leveraging a Constant-curvature Motion Constraint in Radar Odometry
What Goes Around: Leveraging a Constant-curvature Motion Constraint in Radar Odometry Open
This paper presents a method that leverages vehicle motion constraints to refine data associations in a point-based radar odometry system. By using the strong prior on how a non-holonomic robot is constrained to move smoothly through its e…
View article: Depth-SIMS: Semi-Parametric Image and Depth Synthesis
Depth-SIMS: Semi-Parametric Image and Depth Synthesis Open
In this paper we present a compositing image synthesis method that generates RGB canvases with well aligned segmentation maps and sparse depth maps, coupled with an in-painting network that transforms the RGB canvases into high quality RGB…
View article: Fast-MbyM: Leveraging Translational Invariance of the Fourier Transform for Efficient and Accurate Radar Odometry
Fast-MbyM: Leveraging Translational Invariance of the Fourier Transform for Efficient and Accurate Radar Odometry Open
Masking By Moving (MByM), provides robust and accurate radar odometry measurements through an exhaustive correlative search across discretised pose candidates. However, this dense search creates a significant computational bottleneck which…
View article: Contrastive Learning for Unsupervised Radar Place Recognition
Contrastive Learning for Unsupervised Radar Place Recognition Open
We learn, in an unsupervised way, an embedding from sequences of radar images that is suitable for solving the place recognition problem with complex radar data. Our method is based on invariant instance feature learning but is tailored fo…
View article: The Oxford Road Boundaries Dataset
The Oxford Road Boundaries Dataset Open
In this paper we present the Oxford Road Boundaries Dataset, designed for training and testing machine-learning-based road-boundary detection and inference approaches. We have hand-annotated two of the 10 km-long forays from the Oxford Rob…
View article: Unsupervised Place Recognition with Deep Embedding Learning over Radar\n Videos
Unsupervised Place Recognition with Deep Embedding Learning over Radar\n Videos Open
We learn, in an unsupervised way, an embedding from sequences of radar images\nthat is suitable for solving place recognition problem using complex radar\ndata. We experiment on 280 km of data and show performance exceeding\nstate-of-the-a…
View article: Unsupervised Place Recognition with Deep Embedding Learning over Radar Videos
Unsupervised Place Recognition with Deep Embedding Learning over Radar Videos Open
We learn, in an unsupervised way, an embedding from sequences of radar images that is suitable for solving place recognition problem using complex radar data. We experiment on 280 km of data and show performance exceeding state-of-the-art …
View article: Fool Me Once: Robust Selective Segmentation via Out-of-Distribution Detection with Contrastive Learning
Fool Me Once: Robust Selective Segmentation via Out-of-Distribution Detection with Contrastive Learning Open
In this work, we train a network to simultaneously perform segmentation and pixel-wise Out-of-Distribution (OoD) detection, such that the segmentation of unknown regions of scenes can be rejected. This is made possible by leveraging an OoD…
View article: Fool Me Once: Robust Selective Segmentation via Out-of-Distribution\n Detection with Contrastive Learning
Fool Me Once: Robust Selective Segmentation via Out-of-Distribution\n Detection with Contrastive Learning Open
In this work, we train a network to simultaneously perform segmentation and\npixel-wise Out-of-Distribution (OoD) detection, such that the segmentation of\nunknown regions of scenes can be rejected. This is made possible by leveraging\nan …
View article: On the Road: Route Proposal from Radar Self-Supervised by Fuzzy LiDAR Traversability
On the Road: Route Proposal from Radar Self-Supervised by Fuzzy LiDAR Traversability Open
This is motivated by a requirement for robust, autonomy-enabling scene understanding in unknown environments. In the method proposed in this paper, discriminative machine-learning approaches are applied to infer traversability and predict …
View article: Sense–Assess–eXplain (SAX): Building Trust in Autonomous Vehicles in Challenging Real-World Driving Scenarios
Sense–Assess–eXplain (SAX): Building Trust in Autonomous Vehicles in Challenging Real-World Driving Scenarios Open
This paper discusses ongoing work in demonstrating research in mobile autonomy in challenging driving scenarios. In our approach, we address fundamental technical issues to overcome critical barriers to assurance and regulation for large-s…
View article: RSS-Net: Weakly-Supervised Multi-Class Semantic Segmentation with FMCW Radar
RSS-Net: Weakly-Supervised Multi-Class Semantic Segmentation with FMCW Radar Open
This paper presents an efficient annotation procedure and an application thereof to end-to-end, rich semantic segmentation of the sensed environment using FMCW scanning radar. We advocate radar over the traditional sensors used for this ta…