Dr Oscar Mendez Maldonado

This work addresses 3D human pose reconstruction in single images. We present a method that combines Forward Kinematics (FK) with neural networks to ensure a fast and valid prediction of 3D pose. Pose is represented as a hierarchical tree/graph with nodes corresponding to human joints that model their physical limits. Given a 2D detection of keypoints in the image, we lift the skeleton to 3D using neural networks to predict both the joint rotations and bone lengths. These predictions are then combined with skeletal constraints using an FK layer implemented as a network layer in Py-Torch. The result is a fast and accurate approach to the estimation of 3D skeletal pose. Through quantitative and qualitative evaluation , we demonstrate the method is significantly more accurate than MediaPipe in terms of both per joint positional error and visual appearance. Furthermore, we demonstrate generalization over different datasets and sign languages. The implementation in PyTorch runs at between 100-200 milliseconds per image (including CNN detection) using CPU only. Index Terms— 3D pose estimation, hand and body reconstruction .

The use of human-level semantic information to aid robotic tasks has recently become an important area for both Computer Vision and Robotics. This has been enabled by advances in Deep Learning that allow consistent and robust semantic understanding. Leveraging this semantic vision of the world has allowed human-level understanding to naturally emerge from many different approaches. Particularly, the use of semantic information to aid in localisation and reconstruction has been at the forefront of both fields. Like robots, humans also require the ability to localise within a structure. To aid this, humans have designed highlevel semantic maps of our structures called floorplans. We are extremely good at localising in them, even with limited access to the depth information used by robots. This is because we focus on the distribution of semantic elements, rather than geometric ones. Evidence of this is that humans are normally able to localise in a floorplan that has not been scaled properly. In order to grant this ability to robots, it is necessary to use localisation approaches that leverage the same semantic information humans use. In this paper, we present a novel method for semantically enabled global localisation. Our approach relies on the semantic labels present in the floorplan. Deep Learning is leveraged to extract semantic labels from RGB images, which are compared to the floorplan for localisation. While our approach is able to use range measurements if available, we demonstrate that they are unnecessary as we can achieve results comparable to state-of-the-art without them.

This paper presents a new information source for supporting robot localisation: material composition. The proposed method complements the existing visual, structural, and semantic cues utilized in the literature. However, it has a distinct advantage in its ability to differentiate structurally [23], visually [25] or categorically [1] similar objects such as different doors, by using Raman spectrometers. Such devices can identify the material of objects it probes through the bonds between the material’s molecules. Unlike similar sensors, such as mass spectroscopy, it does so without damaging the material or environment. In addition to introducing the first material-based localisation algorithm, this paper supports the future growth of the field by presenting a gazebo plugin for Raman spectrometers, material sensing demonstrations, as well as the first-ever localisation data-set with benchmarks for material-based localisation. This benchmarking shows that the proposed technique results in a significant improvement over current state-of-the-art localisation techniques, achieving 16% more accurate localisation than the leading baseline. The code and dataset will be released at: https://github.com/ThirgoodC/RaSpectLoc

We approach instantaneous mapping, converting images to a top-down view of the world, as a translation problem. We show how a novel form of transformer network can be used to map from images and video directly to an overhead map or bird's-eye-view (BEV) of the world, in a single end-to-end network. We assume a 1-1 correspondence between a vertical scanline in the image, and rays passing through the camera location in an overhead map. This lets us formulate map generation from an image as a set of sequence-to-sequence translations. This constrained formulation , based upon a strong physical grounding of the problem, leads to a restricted transformer network that is convolutional in the horizontal direction only. The structure allows us to make efficient use of data when training, and obtains state-of-the-art results for instantaneous mapping of three large-scale datasets, including a 15% and 30% relative gain against existing best performing methods on the nuScenes and Argoverse datasets, respectively.

Hand pose estimation from a single image has many applications. However, approaches to full 3D body pose estimation are typically trained on day-to-day activities or actions. As such, detailed hand-to-hand interactions are poorly represented, especially during motion. We see this in the failure cases of techniques such as OpenPose [6] or MediaPipe[30]. However, accurate hand pose estimation is crucial for many applications where the global body motion is less important than accurate hand pose estimation. This paper addresses the problem of 3D hand pose estimation from monocular images or sequences. We present a novel end-to-end framework for 3D hand regression that employs diffusion models that have shown excellent ability to capture the distribution of data for generative purposes. Moreover, we enforce kinematic constraints to ensure realistic poses are generated by incorporating an explicit forward kinematic layer as part of the network. The proposed model provides state-of-the-art performance when lifting a 2D single-hand image to 3D. However, when sequence data is available, we add a Transformer module over a temporal window of consecutive frames to refine the results, overcoming jittering and further increasing accuracy. The method is quantitatively and qualitatively evaluated showing state-of-the-art robustness, generalization, and accuracy on several different datasets.

Graph convolutional networks (GCNs) enable end-to-end learning on graph structured data. However, many works assume a given graph structure. When the input graph is noisy or unavailable, one approach is to construct or learn a latent graph structure. These methods typically fix the choice of node degree for the entire graph, which is suboptimal. Instead, we propose a novel end-to-end differentiable graph generator which builds graph topologies where each node selects both its neighborhood and its size. Our module can be readily integrated into existing pipelines involving graph convolution operations, replacing the predetermined or existing adjacency matrix with one that is learned, and optimized, as part of the general objective. As such it is applicable to any GCN. We integrate our module into trajectory prediction, point cloud classification and node classification pipelines resulting in improved accuracy over other structure-learning methods across a wide range of datasets and GCN backbones. We will release the code.

How many times does a human have to drive through the same area to become familiar with it? To begin with, we might first build a mental model of our surroundings. Upon revisiting this area, we can use this model to extrapolate to new unseen locations and imagine their appearance. Based on this, we propose an approach where an agent is capable of modelling new environments after a single visitation. To this end, we introduce “Deep Imagination”, a combination of classical Visual-based Monte Carlo Localisation and deep learning. By making use of a feature embedded 3D map, the system can “imagine” the view from any novel location. These “imagined” views are contrasted with the current observation in order to estimate the agent’s current location. In order to build the embedded map, we train a deep Siamese Fully Convolutional U-Net to perform dense feature extraction. By training these features to be generic, no additional training or fine tuning is required to adapt to new environments. Our results demonstrate the generality and transfer capability of our learnt dense features by training and evaluating on multiple datasets. Additionally, we include several visualizations of the feature representations and resulting 3D maps, as well as their application to localisation.

How many times does a human have to drive through the same area to become familiar with it? To begin with, we might first build a mental model of our surroundings. Upon revisiting this area, we can use this model to extrapolate to new unseen locations and imagine their appearance. Based on this, we propose an approach where an agent is capable of modelling new environments after a single visitation. To this end, we introduce "Deep Imagination", a combination of classical Visual-based Monte Carlo Localisation and deep learning. By making use of a feature embedded 3D map, the system can "imagine" the view from any novel location. These "imagined" views are contrasted with the current observation in order to estimate the agent's current location. In order to build the embedded map, we train a deep Siamese Fully Convolutional U-Net to perform dense feature extraction. By training these features to be generic, no additional training or fine tuning is required to adapt to new environments. Our results demonstrate the generality and transfer capability of our learnt dense features by training and evaluating on multiple datasets. Additionally, we include several visualizations of the feature representations and resulting 3D maps, as well as their application to localisation.

Constructing Birds-Eye-View (BEV) maps from monocular images is typically a complex multi-stage process involving the separate vision tasks of ground plane estimation, road segmentation and 3D object detection. However, recent approaches have adopted end-to-end solutions which warp image-based features from the image-plane to BEV while implicitly taking account of camera geometry. In this work, we show how such instantaneous BEV estimation of a scene can be learnt, and a better state estimation of the world can be achieved by incorporating temporal information. Our model learns a representation from monocular video through factorised 3D convolutions and uses this to estimate a BEV occupancy grid of the final frame. We achieve state-of-the-art results for BEV estimation from monocular images, and establish a new benchmark for single-scene BEV estimation from monocular video.

Attention is an important component of modern deep learning. However, less emphasis has been put on its inverse: ignoring distraction. Our daily lives require us to explicitly avoid giving attention to salient visual features that confound the task we are trying to accomplish. This visual prioritisation allows us to concentrate on important tasks while ignoring visual distractors.In this work, we introduce Neural Blindness, which gives an agent the ability to completely ignore objects or classes that are deemed distractors. More explicitly, we aim to render a neural network completely incapable of representing specific chosen classes in its latent space. In a very real sense, this makes the network "blind" to certain classes, allowing and agent to focus on what is important for a given task, and demonstrates how this can be used to improve localisation.

— Robots need to be able to work in multiple different environments. Even when performing similar tasks, different behaviour should be deployed to best fit the current environment. In this paper, We propose a new approach to navigation, where it is treated as a multi-task learning problem. This enables the robot to learn to behave differently in visual navigation tasks for different environments while also learning shared expertise across environments. We evaluated our approach in both simulated environments as well as real-world data. Our method allows our system to converge with a 26% reduction in training time, while also increasing accuracy.

Sensor-based remote health monitoring is used in industrial, urban and healthcare settings to monitor ongoing operation of equipment and human health. An important aim is to intervene early if anomalous events or adverse health is detected. In the wild, these anomaly detection approaches are challenged by noise, label scarcity, high dimensionality, explainability and wide variability in operating environments. The Contextual Matrix Profile (CMP) is a configurable 2-dimensional version of the Matrix Profile (MP) that uses the distance matrix of all subsequences of a time series to discover patterns and anomalies. The CMP is shown to enhance the effectiveness of the MP and other SOTA methods at detecting, visualising and interpreting true anomalies in noisy real world data from different domains. It excels at zooming out and identifying temporal patterns at configurable time scales. However, the CMP does not address cross-sensor information, and cannot scale to high dimensional data. We propose a novel, self-supervised graph- based approach for temporal anomaly detection that works on context graphs generated from the CMP distance matrix. The learned graph embeddings encode the anomalous nature of a time context. In addition, we evaluate other graph outlier algorithms for the same task. Given our pipeline is modular, graph construction, generation of graph embeddings, and pattern recognition logic can all be chosen based on the specific pattern detection application.We verified the effectiveness of graph-based anomaly detection and compared it with the CMP and 3 state-of-the art methods on two real-world healthcare datasets with different anomalies. Our proposed method demonstrated better recall, alert rate and generalisability.

Accurate extrinsic sensor calibration is essential for both autonomous vehicles and robots. Traditionally this is an involved process requiring calibration targets, known fiducial markers and is generally performed in a lab. Moreover, even a small change in the sensor layout requires recalibration. With the anticipated arrival of consumer autonomous vehicles, there is demand for a system which can do this automatically, after deployment and without specialist human expertise. To solve these limitations, we propose a flexible framework which can estimate extrinsic parameters without an explicit calibration stage, even for sensors with unknown scale. Our first contribution builds upon standard hand-eye calibration by jointly recovering scale. Our second contribution is that our system is made robust to imperfect and degenerate sensor data, by collecting independent sets of poses and automatically selecting those which are most ideal. We show that our approach’s robustness is essential for the target scenario. Unlike previous approaches, ours runs in real time and constantly estimates the extrinsic transform. For both an ideal experimental setup and a real use case, comparison against these approaches shows that we outperform the state-of-the-art. Furthermore, we demonstrate that the recovered scale may be applied to the full trajectory, circumventing the need for scale estimation via sensor fusion.

— Visual Odometry (VO) is used in many applications including robotics and autonomous systems. However, traditional approaches based on feature matching are compu-tationally expensive and do not directly address failure cases, instead relying on heuristic methods to detect failure. In this work, we propose a deep learning-based VO model to efficiently estimate 6-DoF poses, as well as a confidence model for these estimates. We utilise a CNN-RNN hybrid model to learn feature representations from image sequences. We then employ a Mixture Density Network (MDN) which estimates camera motion as a mixture of Gaussians, based on the extracted spatio-temporal representations. Our model uses pose labels as a source of supervision, but derives uncertainties in an unsupervised manner. We evaluate the proposed model on the KITTI and nuScenes datasets and report extensive quantitative and qualitative results to analyse the performance of both pose and uncertainty estimation. Our experiments show that the proposed model exceeds state-of-the-art performance in addition to detecting failure cases using the predicted pose uncertainty.

We approach instantaneous mapping, converting images to a top-down view of the world, as a translation problem. We show how a novel form of transformer network can be used to map from images and video directly to an overhead map or bird's-eye-view (BEV) of the world, in a single end-to-end network. We assume a 1-1 correspondence between a vertical scanline in the image, and rays passing through the camera location in an overhead map. This lets us formulate map generation from an image as a set of sequence-to-sequence translations. Posing the problem as translation allows the network to use the context of the image when interpreting the role of each pixel. This constrained formulation, based upon a strong physical grounding of the problem, leads to a restricted transformer network that is convolutional in the horizontal direction only. The structure allows us to make efficient use of data when training, and obtains state-of-the-art results for instantaneous mapping of three large-scale datasets, including a 15% and 30% relative gain against existing best performing methods on the nuScenes and Argoverse datasets, respectively.

— Visual Odometry (VO) estimation is an important source of information for vehicle state estimation and autonomous driving. Recently, deep learning based approaches have begun to appear in the literature. However, in the context of driving, single sensor based approaches are often prone to failure because of degraded image quality due to environmental factors, camera placement, etc. To address this issue, we propose a deep sensor fusion framework which estimates vehicle motion using both pose and uncertainty estimations from multiple on-board cameras. We extract spatio-temporal feature representations from a set of consecutive images using a hybrid CNN-RNN model. We then utilise a Mixture Density Network (MDN) to estimate the 6-DoF pose as a mixture of distributions and a fusion module to estimate the final pose using MDN outputs from multi-cameras. We evaluate our approach on the publicly available, large scale autonomous vehicle dataset, nuScenes. The results show that the proposed fusion approach surpasses the state-of-the-art, and provides robust estimates and accurate trajectories compared to individual camera-based estimations.

In the context of self-driving vehicles there is strong competition between approaches based on visual localisa-tion 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.

—The ability to produce large-scale maps for navigation , path planning and other tasks is a crucial step for autonomous agents, but has always been challenging. In this work, we introduce BEV-SLAM, a novel type of graph-based SLAM that aligns semantically-segmented Bird's Eye View (BEV) predictions from monocular cameras. We introduce a novel form of occlusion reasoning into BEV estimation and demonstrate its importance to aid spatial aggregation of BEV predictions. The result is a versatile SLAM system that can operate across arbitrary multi-camera configurations and can be seamlessly integrated with other sensors. We show that the use of multiple cameras significantly increases performance, and achieves lower relative error than high-performance GPS. The resulting system is able to create large, dense, globally-consistent world maps from monocular cameras mounted around an ego vehicle. The maps are metric and correctly-scaled, making them suitable for downstream navigation tasks.

Motion estimation approaches typically employ sensor fusion techniques, such as the Kalman Filter, to handle individual sensor failures. More recently, deep learning-based fusion approaches have been proposed, increasing the performance and requiring less model-specific implementations. However, current deep fusion approaches often assume that sensors are synchronised, which is not always practical, especially for low-cost hardware. To address this limitation, in this work, we propose AFT-VO, a novel transformer-based sensor fusion architecture to estimate VO from multiple sensors. Our framework combines predictions from asynchronous multi-view cameras and accounts for the time discrepancies of measurements coming from different sources. Our approach first employs a Mixture Density Network (MDN) to estimate the probability distributions of the 6-DoF poses for every camera in the system. Then a novel transformer-based fusion module, AFT-VO, is introduced, which combines these asynchronous pose estimations, along with their confidences. More specifically, we introduce Discretiser and Source Encoding techniques which enable the fusion of multi-source asynchronous signals. We evaluate our approach on the popular nuScenes and KITTI datasets. Our experiments demonstrate that multi-view fusion for VO estimation provides robust and accurate trajectories, outperforming the state of the art in both challenging weather and lighting conditions.

Robots need to be able to work in multiple different environments. Even when performing similar tasks, different behaviour should be deployed to best fit the current environment. In this paper, We propose a new approach to navigation, where it is treated as a multi-task learning problem. This enables the robot to learn to behave differently in visual navigation tasks for different environments while also learning shared expertise across environments. We evaluated our approach in both simulated environments as well as real-world data. Our method allows our system to converge with a 26% reduction in training time, while also increasing accuracy.

Estimating a semantically segmented bird's-eye-view (BEV) map from a single image has become a popular technique for autonomous control and navigation. However, they show an increase in localization error with distance from the camera. While such an increase in error is entirely expected – localization is harder at distance – much of the drop in performance can be attributed to the cues used by current texture-based models, in particular, they make heavy use of object-ground intersections (such as shadows) [9], which become increasingly sparse and uncertain for distant objects. In this work, we address these shortcomings in BEV-mapping by learning the spatial relationship between objects in a scene. We propose a graph neural network which predicts BEV objects from a monocular image by spatially reasoning about an object within the context of other objects. Our approach sets a new state-of-the-art in BEV estimation from monocular images across three large-scale datasets, including a 50% relative improvement for objects on nuScenes.

In this work, we introduce a new perspective for learning transferable content in multi-task imitation learning. Humans are able to transfer skills and knowledge. If we can cycle to work and drive to the store, we can also cycle to the store and drive to work. We take inspiration from this and hypothesize the latent memory of a policy network can be disentangled into two partitions. These contain either the knowledge of the environmental context for the task or the generalizable skill needed to solve the task. This allows improved training efficiency and better generalization over previously unseen combinations of skills in the same environment, and the same task in unseen environments. We used the proposed approach to train a disentangled agent for two different multi-task IL environments. In both cases we out-performed the SOTA by 30% in task success rate. We also demonstrated this for navigation on a real robot.

How does a person work out their location using a floorplan? It is probably safe to say that we do not explicitly measure depths to every visible surface and try to match them against different pose estimates in the floorplan. And yet, this is exactly how most robotic scan-matching algorithms operate. Similarly, we do not extrude the 2D geometry present in the floorplan into 3D and try to align it to the real-world. And yet, this is how most vision-based approaches localise. Humans do the exact opposite. Instead of depth, we use high level semantic cues. Instead of extruding the floorplan up into the third dimension, we collapse the 3D world into a 2D representation. Evidence of this is that many of the floorplans we use in everyday life are not accurate, opting instead for high levels of discriminative landmarks. In this work, we use this insight to present a global localisation approach that relies solely on the semantic labels present in the floorplan and extracted from RGB images. While our approach is able to use range measurements if available, we demonstrate that they are unnecessary as we can achieve results comparable to state-of-the-art without them.

Reconstruction of 3D environments is a problem that has been widely addressed in the literature. While many approaches exist to perform reconstruction, few of them take an active role in deciding where the next observations should come from. Furthermore, the problem of travelling from the camera’s current position to the next, known as pathplanning, usually focuses on minimising path length. This approach is ill-suited for reconstruction applications, where learning about the environment is more valuable than speed of traversal. We present a novel Scenic Route Planner that selects paths which maximise information gain, both in terms of total map coverage and reconstruction accuracy. We also introduce a new type of collaborative behaviour into the planning stage called opportunistic collaboration, which allows sensors to switch between acting as independent Structure from Motion (SfM) agents or as a variable baseline stereo pair. We show that Scenic Planning enables similar performance to state-of-the-art batch approaches using less than 0.00027% of the possible stereo pairs (3% of the views). Comparison against length-based pathplanning approaches show that our approach produces more complete and more accurate maps with fewer frames. Finally, we demonstrate the Scenic Pathplanner’s ability to generalise to live scenarios by mounting cameras on autonomous ground-based sensor platforms and exploring an environment.

Most 3D reconstruction approaches passively optimise over all data, exhaustively matching pairs, rather than actively selecting data to process. This is costly both in terms of time and computer resources, and quickly becomes intractable for large datasets. This work proposes an approach to intelligently filter large amounts of data for 3D reconstructions of unknown scenes using monocular cameras. Our contributions are twofold: First, we present a novel approach to efficiently optimise the Next-Best View ( NBV ) in terms of accuracy and coverage using partial scene geometry. Second, we extend this to intelligently selecting stereo pairs by jointly optimising the baseline and vergence to find the NBV ’s best stereo pair to perform reconstruction. Both contributions are extremely efficient, taking 0.8ms and 0.3ms per pose, respectively. Experimental evaluation shows that the proposed method allows efficient selection of stereo pairs for reconstruction, such that a dense model can be obtained with only a small number of images. Once a complete model has been obtained, the remaining computational budget is used to intelligently refine areas of uncertainty, achieving results comparable to state-of-the-art batch approaches on the Middlebury dataset, using as little as 3.8% of the views.