Jonathan Kelly

Philippe Nadeau, Matthew Giamou, Jonathan Kelly

To operate safely and efficiently alongside human workers, collaborative robots (cobots) require the ability to quickly understand the dynamics of manipulated objects. However, traditional methods for estimating the full set of inertial parameters rely on motions that are necessarily fast and unsafe (to achieve a sufficient signal-to-noise ratio). In this work, we take an alternative approach: by combining visual and force-torque measurements, we develop an inertial parameter identification algorithm that requires slow or 'stop-and-go' motions only, and hence is ideally tailored for use around humans. Our technique, called Homogeneous Part Segmentation (HPS), leverages the observation that man-made objects are often composed of distinct, homogeneous parts. We combine a surface-based point clustering method with a volumetric shape segmentation algorithm to quickly produce a part-level segmentation of a manipulated object; the segmented representation is then used by HPS to accurately estimate the object's inertial parameters. To benchmark our algorithm, we create and utilize a novel dataset consisting of realistic meshes, segmented point clouds, and inertial parameters for 20 common workshop tools. Finally, we demonstrate the real-world performance and accuracy of HPS by performing an intricate 'hammer balancing act' autonomously and online with a low-cost collaborative robotic arm. Our code and dataset are open source and freely available.

Emmett Wise, Pushyami Kaveti, Qilong Chen et al.

Automatic extrinsic sensor calibration is a fundamental problem for multi-sensor platforms. Reliable and general-purpose solutions should be computationally efficient, require few assumptions about the structure of the sensing environment, and demand little effort from human operators. In this work, we introduce a fast and certifiably globally optimal algorithm for solving a generalized formulation of the robot-world and hand-eye calibration (RWHEC) problem. The formulation of RWHEC presented is "generalized" in that it supports the simultaneous estimation of multiple sensor and target poses, and permits the use of monocular cameras that, alone, are unable to measure the scale of their environments. In addition to demonstrating our method's superior performance over existing solutions through extensive simulated and real experiments, we derive novel identifiability criteria and establish a priori guarantees of global optimality for problem instances with bounded measurement errors. As part of our analysis, we propose a new constraint qualification for nonlinear programs with redundant constraints; this constraint qualification is of independent interest for establishing the exactness of SDP relaxations of QCQPs that have been tightened through the addition of redundant constraints. Finally, we provide a free and open-source implementation of our algorithms and experiments.

Ashkan Mirzaei, Tristan Aumentado-Armstrong, Konstantinos G. Derpanis et al.

Neural Radiance Fields (NeRFs) have emerged as a popular approach for novel view synthesis. While NeRFs are quickly being adapted for a wider set of applications, intuitively editing NeRF scenes is still an open challenge. One important editing task is the removal of unwanted objects from a 3D scene, such that the replaced region is visually plausible and consistent with its context. We refer to this task as 3D inpainting. In 3D, solutions must be both consistent across multiple views and geometrically valid. In this paper, we propose a novel 3D inpainting method that addresses these challenges. Given a small set of posed images and sparse annotations in a single input image, our framework first rapidly obtains a 3D segmentation mask for a target object. Using the mask, a perceptual optimizationbased approach is then introduced that leverages learned 2D image inpainters, distilling their information into 3D space, while ensuring view consistency. We also address the lack of a diverse benchmark for evaluating 3D scene inpainting methods by introducing a dataset comprised of challenging real-world scenes. In particular, our dataset contains views of the same scene with and without a target object, enabling more principled benchmarking of the 3D inpainting task. We first demonstrate the superiority of our approach on multiview segmentation, comparing to NeRFbased methods and 2D segmentation approaches. We then evaluate on the task of 3D inpainting, establishing state-ofthe-art performance against other NeRF manipulation algorithms, as well as a strong 2D image inpainter baseline. Project Page: https://spinnerf3d.github.io

Phone Thiha Kyaw, Jonathan Kelly

Informed sampling techniques accelerate sampling-based motion planners by focusing the search on promising regions of the state space, yet most existing methods rely on Euclidean heuristics that become inadmissible under configuration-dependent Riemannian metrics. While scalar eigenvalue bounds restore admissibility by uniformly scaling the Euclidean distance, they discard the directional structure of the metric, producing overly conservative informed sets. We propose a matrix-valued admissible heuristic that exploits the Loewner order on symmetric positive definite matrices to compute the tightest constant lower bound on the metric tensor while preserving its full directional structure. The Cholesky factorization of this bound defines a linear map to an isotropic Euclidean space in which the Riemannian informed set reduces to a standard prolate hyperspheroid, enabling direct, rejection-free sampling using existing algorithms. Experiments on manipulation tasks with a 6-DoF UR5, 7-DoF Franka, and 14-DoF PR2 under three distinct Riemannian metrics show that our heuristic produces consistently tighter informed sets than both the Euclidean and scalar eigenvalue bounds, accelerating convergence across multiple state-of-the-art asymptotically optimal planners.

Ashkan Mirzaei, Tristan Aumentado-Armstrong, Marcus A. Brubaker et al.

Denoising diffusion models have enabled high-quality image generation and editing. We present a method to localize the desired edit region implicit in a text instruction. We leverage InstructPix2Pix (IP2P) and identify the discrepancy between IP2P predictions with and without the instruction. This discrepancy is referred to as the relevance map. The relevance map conveys the importance of changing each pixel to achieve the edits, and is used to to guide the modifications. This guidance ensures that the irrelevant pixels remain unchanged. Relevance maps are further used to enhance the quality of text-guided editing of 3D scenes in the form of neural radiance fields. A field is trained on relevance maps of training views, denoted as the relevance field, defining the 3D region within which modifications should be made. We perform iterative updates on the training views guided by rendered relevance maps from the relevance field. Our method achieves state-of-the-art performance on both image and NeRF editing tasks. Project page: https://ashmrz.github.io/WatchYourSteps/

Ashkan Mirzaei, Tristan Aumentado-Armstrong, Marcus A. Brubaker et al.

The popularity of Neural Radiance Fields (NeRFs) for view synthesis has led to a desire for NeRF editing tools. Here, we focus on inpainting regions in a view-consistent and controllable manner. In addition to the typical NeRF inputs and masks delineating the unwanted region in each view, we require only a single inpainted view of the scene, i.e., a reference view. We use monocular depth estimators to back-project the inpainted view to the correct 3D positions. Then, via a novel rendering technique, a bilateral solver can construct view-dependent effects in non-reference views, making the inpainted region appear consistent from any view. For non-reference disoccluded regions, which cannot be supervised by the single reference view, we devise a method based on image inpainters to guide both the geometry and appearance. Our approach shows superior performance to NeRF inpainting baselines, with the additional advantage that a user can control the generated scene via a single inpainted image. Project page: https://ashmrz.github.io/reference-guided-3d

Ashkan Mirzaei, Yash Kant, Jonathan Kelly et al.

Obtaining 3D object representations is important for creating photo-realistic simulations and for collecting AR and VR assets. Neural fields have shown their effectiveness in learning a continuous volumetric representation of a scene from 2D images, but acquiring object representations from these models with weak supervision remains an open challenge. In this paper we introduce LaTeRF, a method for extracting an object of interest from a scene given 2D images of the entire scene, known camera poses, a natural language description of the object, and a set of point-labels of object and non-object points in the input images. To faithfully extract the object from the scene, LaTeRF extends the NeRF formulation with an additional `objectness' probability at each 3D point. Additionally, we leverage the rich latent space of a pre-trained CLIP model combined with our differentiable object renderer, to inpaint the occluded parts of the object. We demonstrate high-fidelity object extraction on both synthetic and real-world datasets and justify our design choices through an extensive ablation study.

Phone Thiha Kyaw, Jonathan Kelly

In many robot motion planning problems, task objectives and physical constraints induce non-Euclidean geometry on the configuration space, yet many planners operate using Euclidean distances that ignore this structure. We address the problem of planning collision-free motions that minimize length under configuration-dependent Riemannian metrics, corresponding to geodesics on the configuration manifold. Conventional numerical methods for computing such paths do not scale well to high-dimensional systems, while sampling-based planners trade scalability for geometric fidelity. To bridge this gap, we propose a sampling-based motion planning framework that operates directly on Riemannian manifolds. We introduce a computationally efficient midpoint-based approximation of the Riemannian geodesic distance and prove that it matches the true Riemannian distance with third-order accuracy. Building on this approximation, we design a local planner that traces the manifold using first-order retractions guided by Riemannian natural gradients. Experiments on a two-link planar arm and a 7-DoF Franka manipulator under a kinetic-energy metric, as well as on rigid-body planning in $\mathrm{SE}(2)$ with non-holonomic motion constraints, demonstrate that our approach consistently produces lower-cost trajectories than Euclidean-based planners and classical numerical geodesic-solver baselines.

Tristan Aumentado-Armstrong, Ashkan Mirzaei, Marcus A. Brubaker et al.

Neural Radiance Fields (NeRFs) have proven to be powerful 3D representations, capable of high quality novel view synthesis of complex scenes. While NeRFs have been applied to graphics, vision, and robotics, problems with slow rendering speed and characteristic visual artifacts prevent adoption in many use cases. In this work, we investigate combining an autoencoder (AE) with a NeRF, in which latent features (instead of colours) are rendered and then convolutionally decoded. The resulting latent-space NeRF can produce novel views with higher quality than standard colour-space NeRFs, as the AE can correct certain visual artifacts, while rendering over three times faster. Our work is orthogonal to other techniques for improving NeRF efficiency. Further, we can control the tradeoff between efficiency and image quality by shrinking the AE architecture, achieving over 13 times faster rendering with only a small drop in performance. We hope that our approach can form the basis of an efficient, yet high-fidelity, 3D scene representation for downstream tasks, especially when retaining differentiability is useful, as in many robotics scenarios requiring continual learning.

Trevor Ablett, Bryan Chan, Jonathan Kelly

Adversarial imitation learning (AIL) has become a popular alternative to supervised imitation learning that reduces the distribution shift suffered by the latter. However, AIL requires effective exploration during an online reinforcement learning phase. In this work, we show that the standard, naive approach to exploration can manifest as a suboptimal local maximum if a policy learned with AIL sufficiently matches the expert distribution without fully learning the desired task. This can be particularly catastrophic for manipulation tasks, where the difference between an expert and a non-expert state-action pair is often subtle. We present Learning from Guided Play (LfGP), a framework in which we leverage expert demonstrations of multiple exploratory, auxiliary tasks in addition to a main task. The addition of these auxiliary tasks forces the agent to explore states and actions that standard AIL may learn to ignore. Additionally, this particular formulation allows for the reusability of expert data between main tasks. Our experimental results in a challenging multitask robotic manipulation domain indicate that LfGP significantly outperforms both AIL and behaviour cloning, while also being more expert sample efficient than these baselines. To explain this performance gap, we provide further analysis of a toy problem that highlights the coupling between a local maximum and poor exploration, and also visualize the differences between the learned models from AIL and LfGP.

Oliver Limoyo, Trevor Ablett, Jonathan Kelly

Sequential modelling of high-dimensional data is an important problem that appears in many domains including model-based reinforcement learning and dynamics identification for control. Latent variable models applied to sequential data (i.e., latent dynamics models) have been shown to be a particularly effective probabilistic approach to solve this problem, especially when dealing with images. However, in many application areas (e.g., robotics), information from multiple sensing modalities is available -- existing latent dynamics methods have not yet been extended to effectively make use of such multimodal sequential data. Multimodal sensor streams can be correlated in a useful manner and often contain complementary information across modalities. In this work, we present a self-supervised generative modelling framework to jointly learn a probabilistic latent state representation of multimodal data and the respective dynamics. Using synthetic and real-world datasets from a multimodal robotic planar pushing task, we demonstrate that our approach leads to significant improvements in prediction and representation quality. Furthermore, we compare to the common learning baseline of concatenating each modality in the latent space and show that our principled probabilistic formulation performs better. Finally, despite being fully self-supervised, we demonstrate that our method is nearly as effective as an existing supervised approach that relies on ground truth labels.

Trevor Ablett, Oliver Limoyo, Adam Sigal et al.

Contact-rich tasks continue to present many challenges for robotic manipulation. In this work, we leverage a multimodal visuotactile sensor within the framework of imitation learning (IL) to perform contact-rich tasks that involve relative motion (e.g., slipping and sliding) between the end-effector and the manipulated object. We introduce two algorithmic contributions, tactile force matching and learned mode switching, as complimentary methods for improving IL. Tactile force matching enhances kinesthetic teaching by reading approximate forces during the demonstration and generating an adapted robot trajectory that recreates the recorded forces. Learned mode switching uses IL to couple visual and tactile sensor modes with the learned motion policy, simplifying the transition from reaching to contacting. We perform robotic manipulation experiments on four door-opening tasks with a variety of observation and algorithm configurations to study the utility of multimodal visuotactile sensing and our proposed improvements. Our results show that the inclusion of force matching raises average policy success rates by 62.5%, visuotactile mode switching by 30.3%, and visuotactile data as a policy input by 42.5%, emphasizing the value of see-through tactile sensing for IL, both for data collection to allow force matching, and for policy execution to enable accurate task feedback. Project site: https://papers.starslab.ca/sts-il/

Olivier Lamarre, Jonathan Kelly

In robotic planetary surface exploration, strategic mobility planning is an important task that involves finding candidate long-distance routes on orbital maps and identifying segments with uncertain traversability. Then, expert human operators establish safe, adaptive traverse plans based on the actual navigation difficulties encountered in these uncertain areas. In this paper, we formalize this challenge as a new, risk-averse variant of the Canadian Traveller Problem (CTP) tailored to global planetary mobility. The objective is to find a traverse policy minimizing a conditional value-at-risk (CVaR) criterion, which is a risk measure with an intuitive interpretation. We propose a novel search algorithm that finds exact CVaR-optimal policies. Our approach leverages well-established optimal AND-OR search techniques intended for (risk-agnostic) expectation minimization and extends these methods to the risk-averse domain. We validate our approach through simulated long-distance planetary surface traverses; we employ real orbital maps of the Martian surface to construct problem instances and use terrain maps to express traversal probabilities in uncertain regions. Our results illustrate different adaptive decision-making schemes depending on the level of risk aversion. Additionally, our problem setup allows accounting for traversability correlations between similar areas of the environment. In such a case, we empirically demonstrate how information-seeking detours can mitigate risk.

Andrej Janda, Brandon Wagstaff, Edwin G. Ng et al.

Reducing the quantity of annotations required for supervised training is vital when labels are scarce and costly. This reduction is especially important for semantic segmentation tasks involving 3D datasets that are often significantly smaller and more challenging to annotate than their image-based counterparts. Self-supervised pre-training on large unlabelled datasets is one way to reduce the amount of manual annotations needed. Previous work has focused on pre-training with point cloud data exclusively; this approach often requires two or more registered views. In the present work, we combine image and point cloud modalities, by first learning self-supervised image features and then using these features to train a 3D model. By incorporating image data, which is often included in many 3D datasets, our pre-training method only requires a single scan of a scene. We demonstrate that our pre-training approach, despite using single scans, achieves comparable performance to other multi-scan, point cloud-only methods.

Andrej Janda, Brandon Wagstaff, Edwin G. Ng et al.

Reducing the quantity of annotations required for supervised training is vital when labels are scarce and costly. This reduction is particularly important for semantic segmentation tasks involving 3D datasets, which are often significantly smaller and more challenging to annotate than their image-based counterparts. Self-supervised pre-training on unlabelled data is one way to reduce the amount of manual annotations needed. Previous work has focused on pre-training with point clouds exclusively. While useful, this approach often requires two or more registered views. In the present work, we combine image and point cloud modalities by first learning self-supervised image features and then using these features to train a 3D model. By incorporating image data, which is often included in many 3D datasets, our pre-training method only requires a single scan of a scene and can be applied to cases where localization information is unavailable. We demonstrate that our pre-training approach, despite using single scans, achieves comparable performance to other multi-scan, point cloud-only methods.

Vincenzo Polizzi, David B. Lindell, Jonathan Kelly

Event cameras provide several unique advantages over standard frame-based sensors, including high temporal resolution, low latency, and robustness to extreme lighting. However, existing learning-based approaches for event processing are typically confined to narrow, task-specific silos and lack the ability to generalize across modalities. We address this gap with REALM, a cross-modal framework that learns an RGB and Event Aligned Latent Manifold by projecting event representations into the pretrained latent space of RGB foundation models. Instead of task-specific training, we leverage low-rank adaptation (LoRA) to bridge the modality gap, effectively unlocking the geometric and semantic priors of frozen RGB backbones for asynchronous event streams. We demonstrate that REALM effectively maps events into the ViT-based foundation latent space. Our method allows us to perform downstream tasks like depth estimation and semantic segmentation by simply transferring linear heads trained on the RGB teacher. Most significantly, REALM enables the direct, zero-shot application of complex, frozen image-trained decoders, such as MASt3R, to raw event data. We demonstrate state-of-the-art performance in wide-baseline feature matching, significantly outperforming specialized architectures. Code and models are available upon acceptance.

Vincenzo Polizzi, Marco Cannici, Davide Scaramuzza et al.

Camera relocalization methods range from dense image alignment to direct camera pose regression from a query image. Among these, sparse feature matching stands out as an efficient, versatile, and generally lightweight approach with numerous applications. However, feature-based methods often struggle with significant viewpoint and appearance changes, leading to matching failures and inaccurate pose estimates. To overcome this limitation, we propose a novel approach that leverages a globally sparse yet locally dense 3D representation of 2D features. By tracking and triangulating landmarks over a sequence of frames, we construct a sparse voxel map optimized to render image patch descriptors observed during tracking. Given an initial pose estimate, we first synthesize descriptors from the voxels using volumetric rendering and then perform feature matching to estimate the camera pose. This methodology enables the generation of descriptors for unseen views, enhancing robustness to view changes. We extensively evaluate our method on the 7-Scenes and Cambridge Landmarks datasets. Our results show that our method significantly outperforms existing state-of-the-art feature representation techniques in indoor environments, achieving up to a 39% improvement in median translation error. Additionally, our approach yields comparable results to other methods for outdoor scenarios while maintaining lower memory and computational costs.

Trevor Ablett, Bryan Chan, Jayce Haoran Wang et al.

Common approaches to providing feedback in reinforcement learning are the use of hand-crafted rewards or full-trajectory expert demonstrations. Alternatively, one can use examples of completed tasks, but such an approach can be extremely sample inefficient. We introduce value-penalized auxiliary control from examples (VPACE), an algorithm that significantly improves exploration in example-based control by adding examples of simple auxiliary tasks and an above-success-level value penalty. Across both simulated and real robotic environments, we show that our approach substantially improves learning efficiency for challenging tasks, while maintaining bounded value estimates. Preliminary results also suggest that VPACE may learn more efficiently than the more common approaches of using full trajectories or true sparse rewards. Project site: https://papers.starslab.ca/vpace/.

Trevor Ablett, Bryan Chan, Jonathan Kelly

Effective exploration continues to be a significant challenge that prevents the deployment of reinforcement learning for many physical systems. This is particularly true for systems with continuous and high-dimensional state and action spaces, such as robotic manipulators. The challenge is accentuated in the sparse rewards setting, where the low-level state information required for the design of dense rewards is unavailable. Adversarial imitation learning (AIL) can partially overcome this barrier by leveraging expert-generated demonstrations of optimal behaviour and providing, essentially, a replacement for dense reward information. Unfortunately, the availability of expert demonstrations does not necessarily improve an agent's capability to explore effectively and, as we empirically show, can lead to inefficient or stagnated learning. We present Learning from Guided Play (LfGP), a framework in which we leverage expert demonstrations of, in addition to a main task, multiple auxiliary tasks. Subsequently, a hierarchical model is used to learn each task reward and policy through a modified AIL procedure, in which exploration of all tasks is enforced via a scheduler composing different tasks together. This affords many benefits: learning efficiency is improved for main tasks with challenging bottleneck transitions, expert data becomes reusable between tasks, and transfer learning through the reuse of learned auxiliary task models becomes possible. Our experimental results in a challenging multitask robotic manipulation domain indicate that our method compares favourably to supervised imitation learning and to a state-of-the-art AIL method. Code is available at https://github.com/utiasSTARS/lfgp.

Filip Maric, Matthew Giamou, Soroush Khoubyarian et al.

Inverse kinematics is a fundamental problem for articulated robots: fast and accurate algorithms are needed for translating task-related workspace constraints and goals into feasible joint configurations. In general, inverse kinematics for serial kinematic chains is a difficult nonlinear problem, for which closed form solutions cannot be easily obtained. Therefore, computationally efficient numerical methods that can be adapted to a general class of manipulators are of great importance. % to motion planning and workspace generation tasks. In this paper, we use convex optimization techniques to solve the inverse kinematics problem with joint limit constraints for highly redundant serial kinematic chains with spherical joints in two and three dimensions. This is accomplished through a novel formulation of inverse kinematics as a nearest point problem, and with a fast sum of squares solver that exploits the sparsity of kinematic constraints for serial manipulators. Our method has the advantages of post-hoc certification of global optimality and a runtime that scales polynomialy with the number of degrees of freedom. Additionally, we prove that our convex relaxation leads to a globally optimal solution when certain conditions are met, and demonstrate empirically that these conditions are common and represent many practical instances. Finally, we provide an open source implementation of our algorithm.

Matthew Giamou, Filip Maric, Valentin Peretroukhin et al.

Estimating unknown rotations from noisy measurements is an important step in SfM and other 3D vision tasks. Typically, local optimization methods susceptible to returning suboptimal local minima are used to solve the rotation averaging problem. A new wave of approaches that leverage convex relaxations have provided the first formal guarantees of global optimality for state estimation techniques involving SO(3). However, most of these guarantees are only applicable when the measurement error introduced by noise is within a certain bound that depends on the problem instance's structure. In this paper, we cast rotation averaging as a polynomial optimization problem over unit quaternions to produce the first rotation averaging method that is formally guaranteed to provide a certifiably globally optimal solution for \textit{any} problem instance. This is achieved by formulating and solving a sparse convex sum of squares (SOS) relaxation of the problem. We provide an open source implementation of our algorithm and experiments, demonstrating the benefits of our globally optimal approach.

Yilun Wu, Xintong Du, Rikky Duivenvoorden et al.

In this paper, we introduce the Phoenix drone: the first completely open-source tail-sitter micro aerial vehicle (MAV) platform. The vehicle has a highly versatile, dual-rotor design and is engineered to be low-cost and easily extensible/modifiable. Our open-source release includes all of the design documents, software resources, and simulation tools needed to build and fly a high-performance tail-sitter for research and educational purposes. The drone has been developed for precision flight with a high degree of control authority. Our design methodology included extensive testing and characterization of the aerodynamic properties of the vehicle. The platform incorporates many off-the-shelf components and 3D-printed parts, in order to keep the cost down. Nonetheless, the paper includes results from flight trials which demonstrate that the vehicle is capable of very stable hovering and accurate trajectory tracking. Our hope is that the open-source Phoenix reference design will be useful to both researchers and educators. In particular, the details in this paper and the available open-source materials should enable learners to gain an understanding of aerodynamics, flight control, state estimation, software design, and simulation, while experimenting with a unique aerial robot.

Lee Clement, Jonathan Kelly

Direct visual localization has recently enjoyed a resurgence in popularity with the increasing availability of cheap mobile computing power. The competitive accuracy and robustness of these algorithms compared to state-of-the-art feature-based methods, as well as their natural ability to yield dense maps, makes them an appealing choice for a variety of mobile robotics applications. However, direct methods remain brittle in the face of appearance change due to their underlying assumption of photometric consistency, which is commonly violated in practice. In this paper, we propose to mitigate this problem by training deep convolutional encoder-decoder models to transform images of a scene such that they correspond to a previously-seen canonical appearance. We validate our method in multiple environments and illumination conditions using high-fidelity synthetic RGB-D datasets, and integrate the trained models into a direct visual localization pipeline, yielding improvements in visual odometry (VO) accuracy through time-varying illumination conditions, as well as improved metric relocalization performance under illumination change, where conventional methods normally fail. We further provide a preliminary investigation of transfer learning from synthetic to real environments in a localization context. An open-source implementation of our method using PyTorch is available at https://github.com/utiasSTARS/cat-net.

Valentin Peretroukhin, Lee Clement, Jonathan Kelly

We present a method to incorporate global orientation information from the sun into a visual odometry pipeline using only the existing image stream, where the sun is typically not visible. We leverage recent advances in Bayesian Convolutional Neural Networks to train and implement a sun detection model that infers a three-dimensional sun direction vector from a single RGB image. Crucially, our method also computes a principled uncertainty associated with each prediction, using a Monte Carlo dropout scheme. We incorporate this uncertainty into a sliding window stereo visual odometry pipeline where accurate uncertainty estimates are critical for optimal data fusion. Our Bayesian sun detection model achieves a median error of approximately 12 degrees on the KITTI odometry benchmark training set, and yields improvements of up to 42% in translational ARMSE and 32% in rotational ARMSE compared to standard VO. An open source implementation of our Bayesian CNN sun estimator (Sun-BCNN) using Caffe is available at https://github. com/utiasSTARS/sun-bcnn-vo

Ashkan Mirzaei, Riccardo De Lutio, Seung Wook Kim et al.

Neural reconstruction approaches are rapidly emerging as the preferred representation for 3D scenes, but their limited editability is still posing a challenge. In this work, we propose an approach for 3D scene inpainting -- the task of coherently replacing parts of the reconstructed scene with desired content. Scene inpainting is an inherently ill-posed task as there exist many solutions that plausibly replace the missing content. A good inpainting method should therefore not only enable high-quality synthesis but also a high degree of control. Based on this observation, we focus on enabling explicit control over the inpainted content and leverage a reference image as an efficient means to achieve this goal. Specifically, we introduce RefFusion, a novel 3D inpainting method based on a multi-scale personalization of an image inpainting diffusion model to the given reference view. The personalization effectively adapts the prior distribution to the target scene, resulting in a lower variance of score distillation objective and hence significantly sharper details. Our framework achieves state-of-the-art results for object removal while maintaining high controllability. We further demonstrate the generality of our formulation on other downstream tasks such as object insertion, scene outpainting, and sparse view reconstruction.

Philippe Nadeau, Jonathan Kelly

We introduce a planner designed to guide robot manipulators in stably placing objects within intricate scenes. Our proposed method reverses the traditional approach to object placement: our planner selects contact points first and then determines a placement pose that solicits the selected points. This is instead of sampling poses, identifying contact points, and evaluating pose quality. Our algorithm facilitates stability-aware object placement planning, imposing no restrictions on object shape, convexity, or mass density homogeneity, while avoiding combinatorial computational complexity. Our proposed stability heuristic enables our planner to find a solution about 20 times faster when compared to the same algorithm not making use of the heuristic and eight times faster than a state-of-the-art method that uses the traditional sample-and-evaluate approach. Our proposed planner is also more successful in finding stable placements than the five other benchmarked algorithms. Derived from first principles and validated in ten real robot experiments, our planner offers a general and scalable method to tackle the problem of object placement planning with rigid objects.

Selina Leveugle, Chang Won Lee, Svetlana Stolpner et al.

NASA's forthcoming Lunar Gateway space station, which will be uncrewed most of the time, will need to operate with an unprecedented level of autonomy. One key challenge is enabling the Canadarm3, the Gateway's external robotic system, to detect hazards in its environment using its onboard inspection cameras. This task is complicated by the extreme and variable lighting conditions in space. In this paper, we introduce the visual anomaly detection and localization task for the space domain and establish a benchmark based on a synthetic dataset called ALLO (Anomaly Localization in Lunar Orbit). We show that state-of-the-art visual anomaly detection methods often fail in the space domain, motivating the need for new approaches. To address this, we propose MRAD (Model Reference Anomaly Detection), a statistical algorithm that leverages the known pose of the Canadarm3 and a CAD model of the Gateway to generate reference images of the expected scene appearance. Anomalies are then identified as deviations from this model-generated reference. On the ALLO dataset, MRAD surpasses state-of-the-art anomaly detection algorithms, achieving an AP score of 62.9% at the pixel level and an AUROC score of 75.0% at the image level. Given the low tolerance for risk in space operations and the lack of domain-specific data, we emphasize the need for novel, robust, and accurate anomaly detection methods to handle the challenging visual conditions found in lunar orbit and beyond.

Philippe Nadeau, Miguel Rogel, Ivan Bilić et al.

Stably placing an object in a multi-object scene is a fundamental challenge in robotic manipulation, as placements must be penetration-free, establish precise surface contact, and result in a force equilibrium. To assess stability, existing methods rely on running a simulation engine or resort to heuristic, appearance-based assessments. In contrast, our approach integrates stability directly into the sampling process of a diffusion model. To this end, we query an offline sampling-based planner to gather multi-modal placement labels and train a diffusion model to generate stable placements. The diffusion model is conditioned on scene and object point clouds, and serves as a geometry-aware prior. We leverage the compositional nature of score-based generative models to combine this learned prior with a stability-aware loss, thereby increasing the likelihood of sampling from regions of high stability. Importantly, this strategy requires no additional re-training or fine-tuning, and can be directly applied to off-the-shelf models. We evaluate our method on four benchmark scenes where stability can be accurately computed. Our physics-guided models achieve placements that are 56% more robust to forceful perturbations while reducing runtime by 47% compared to a state-of-the-art geometric method.

Vincenzo Polizzi, Stephen Yang, Quentin Clark et al.

Event cameras are a bio-inspired class of sensors that asynchronously measure per-pixel intensity changes. Under fixed illumination conditions in static or low-motion scenes, rigidly mounted event cameras are unable to generate any events, becoming unsuitable for most computer vision tasks. To address this limitation, recent work has investigated motion-induced event stimulation that often requires complex hardware or additional optical components. In contrast, we introduce a lightweight approach to sustain persistent event generation by employing a simple rotating unbalanced mass to induce periodic vibrational motion. This is combined with a motion-compensation pipeline that removes the injected motion and yields clean, motion-corrected events for downstream perception tasks. We demonstrate our approach with a hardware prototype and evaluate it on real-world captured datasets. Our method reliably recovers motion parameters and improves both image reconstruction and edge detection over event-based sensing without motion induction.

Mia Thomas, Trevor Ablett, Jonathan Kelly

Unmanned aerial vehicles (UAVs) enable operations in remote and hazardous environments, yet the visible-spectrum, camera-based navigation systems often relied upon by UAVs struggle in low-visibility conditions. Thermal cameras, which capture long-wave infrared radiation, are able to function effectively in darkness and smoke, where visible-light cameras fail. This work explores learned cross-spectral (thermal-visible) point features as a means to integrate thermal imagery into established camera-based navigation systems. Existing methods typically train a feature network's detection and description outputs directly, which often focuses training on image regions where thermal and visible-spectrum images exhibit similar appearance. Aiming to more fully utilize the available data, we propose a method to train the feature network on the tasks of matching and registration. We run our feature network on thermal-visible image pairs, then feed the network response into a differentiable registration pipeline. Losses are applied to the matching and registration estimates of this pipeline. Our selected model, trained on the task of matching, achieves a registration error (corner error) below 10 pixels for more than 75% of estimates on the MultiPoint dataset. We further demonstrate that our model can also be used with a classical pipeline for matching and registration.

Chang Won Lee, Selina Leveugle, Svetlana Stolpner et al. · utoronto

Anomaly segmentation is a valuable computer vision task for safety-critical applications that need to be aware of unexpected events. Current state-of-the-art (SOTA) scene-level anomaly segmentation approaches rely on diverse inlier class labels during training, limiting their ability to leverage vast unlabeled datasets and pre-trained vision encoders. These methods may underperform in domains with reduced color diversity and limited object classes. Conversely, existing unsupervised methods struggle with anomaly segmentation with the diverse scenes of less restricted domains. To address these challenges, we introduce FlowCLAS, a novel self-supervised framework that utilizes vision foundation models to extract rich features and employs a normalizing flow network to learn their density distribution. We enhance the model's discriminative power by incorporating Outlier Exposure and contrastive learning in the latent space. FlowCLAS significantly outperforms all existing methods on the ALLO anomaly segmentation benchmark for space robotics and demonstrates competitive results on multiple road anomaly segmentation benchmarks for autonomous driving, including Fishyscapes Lost&Found and Road Anomaly. These results highlight FlowCLAS's effectiveness in addressing the unique challenges of space anomaly segmentation while retaining SOTA performance in the autonomous driving domain without reliance on inlier segmentation labels.

Oliver Limoyo, Jimmy Li, Dmitriy Rivkin et al.

We introduce PhotoBot, a framework for fully automated photo acquisition based on an interplay between high-level human language guidance and a robot photographer. We propose to communicate photography suggestions to the user via reference images that are selected from a curated gallery. We leverage a visual language model (VLM) and an object detector to characterize the reference images via textual descriptions and then use a large language model (LLM) to retrieve relevant reference images based on a user's language query through text-based reasoning. To correspond the reference image and the observed scene, we exploit pre-trained features from a vision transformer capable of capturing semantic similarity across marked appearance variations. Using these features, we compute suggested pose adjustments for an RGB-D camera by solving a perspective-n-point (PnP) problem. We demonstrate our approach using a manipulator equipped with a wrist camera. Our user studies show that photos taken by PhotoBot are often more aesthetically pleasing than those taken by users themselves, as measured by human feedback. We also show that PhotoBot can generalize to other reference sources such as paintings.

Oliver Limoyo, Abhisek Konar, Trevor Ablett et al.

We present placing via picking (PvP), a method to autonomously collect real-world demonstrations for a family of placing tasks in which objects must be manipulated to specific, contact-constrained locations. With PvP, we approach the collection of robotic object placement demonstrations by reversing the grasping process and exploiting the inherent symmetry of the pick and place problems. Specifically, we obtain placing demonstrations from a set of grasp sequences of objects initially located at their target placement locations. Our system can collect hundreds of demonstrations in contact-constrained environments without human intervention using two modules: compliant control for grasping and tactile regrasping. We train a policy directly from visual observations through behavioural cloning, using the autonomously-collected demonstrations. By doing so, the policy can generalize to object placement scenarios outside of the training environment without privileged information (e.g., placing a plate picked up from a table). We validate our approach in home robot scenarios that include dishwasher loading and table setting. Our approach yields robotic placing policies that outperform policies trained with kinesthetic teaching, both in terms of success rate and data efficiency, while requiring no human supervision.

Sean Wu, Nicole Amenta, Jiachen Zhou et al.

Following four successful years in the SAE AutoDrive Challenge Series I, the University of Toronto is participating in the Series II competition to develop a Level 4 autonomous passenger vehicle capable of handling various urban driving scenarios by 2025. Accurate detection of traffic lights and correct identification of their states is essential for safe autonomous operation in cities. Herein, we describe our recently-redesigned traffic light perception system for autonomous vehicles like the University of Toronto's self-driving car, Artemis. Similar to most traffic light perception systems, we rely primarily on camera-based object detectors. We deploy the YOLOv5 detector for bounding box regression and traffic light classification across multiple cameras and fuse the observations. To improve robustness, we incorporate priors from high-definition semantic maps and perform state filtering using hidden Markov models. We demonstrate a multi-camera, real time-capable traffic light perception pipeline that handles complex situations including multiple visible intersections, traffic light variations, temporary occlusion, and flashing light states. To validate our system, we collected and annotated a varied dataset incorporating flashing states and a range of occlusion types. Our results show superior performance in challenging real-world scenarios compared to single-frame, single-camera object detection.

Andrej Janda, Pierre Merriaux, Pierre Olivier et al.

Advances in lidar technology have made the collection of 3D point clouds fast and easy. While most lidar sensors return per-point intensity (or reflectance) values along with range measurements, flash lidar sensors are able to provide information about the shape of the return pulse. The shape of the return waveform is affected by many factors, including the distance that the light pulse travels and the angle of incidence with a surface. Importantly, the shape of the return waveform also depends on the material properties of the reflecting surface. In this paper, we investigate whether the material type or class can be determined from the full-waveform response. First, as a proof of concept, we demonstrate that the extra information about material class, if known accurately, can improve performance on scene understanding tasks such as semantic segmentation. Next, we learn two different full-waveform material classifiers: a random forest classifier and a temporal convolutional neural network (TCN) classifier. We find that, in some cases, material types can be distinguished, and that the TCN generally performs better across a wider range of materials. However, factors such as angle of incidence, material colour, and material similarity may hinder overall performance.

Abhinav Grover, Philippe Nadeau, Christopher Grebe et al.

The ability to perceive object slip via tactile feedback enables humans to accomplish complex manipulation tasks including maintaining a stable grasp. Despite the utility of tactile information for many applications, tactile sensors have yet to be widely deployed in industrial robotics settings; part of the challenge lies in identifying slip and other events from the tactile data stream. In this paper, we present a learning-based method to detect slip using barometric tactile sensors. These sensors have many desirable properties including high durability and reliability, and are built from inexpensive, off-the-shelf components. We train a temporal convolution neural network to detect slip, achieving high detection accuracies while displaying robustness to the speed and direction of the slip motion. Further, we test our detector on two manipulation tasks involving a variety of common objects and demonstrate successful generalization to real-world scenarios not seen during training. We argue that barometric tactile sensing technology, combined with data-driven learning, is suitable for many manipulation tasks such as slip compensation.

Christopher Grebe, Emmett Wise, Jonathan Kelly

Ideally, robots should move in ways that maximize the knowledge gained about the state of both their internal system and the external operating environment. Trajectory design is a challenging problem that has been investigated from a variety of perspectives, ranging from information-theoretic analyses to leaning-based approaches. Recently, observability-based metrics have been proposed to find trajectories that enable rapid and accurate state and parameter estimation. The viability and efficacy of these methods is not yet well understood in the literature. In this paper, we compare two state-of-the-art methods for observability-aware trajectory optimization and seek to add important theoretical clarifications and valuable discussion about their overall effectiveness. For evaluation, we examine the representative task of sensor-to-sensor extrinsic self-calibration using a realistic physics simulator. We also study the sensitivity of these algorithms to changes in the information content of the exteroceptive sensor measurements.

Matthew Giamou, Filip Marić, David M. Rosen et al.

Inverse kinematics (IK) is the problem of finding robot joint configurations that satisfy constraints on the position or pose of one or more end-effectors. For robots with redundant degrees of freedom, there is often an infinite, nonconvex set of solutions. The IK problem is further complicated when collision avoidance constraints are imposed by obstacles in the workspace. In general, closed-form expressions yielding feasible configurations do not exist, motivating the use of numerical solution methods. However, these approaches rely on local optimization of nonconvex problems, often requiring an accurate initialization or numerous re-initializations to converge to a valid solution. In this work, we first formulate inverse kinematics with complex workspace constraints as a convex feasibility problem whose low-rank feasible points provide exact IK solutions. We then present \texttt{CIDGIK} (Convex Iteration for Distance-Geometric Inverse Kinematics), an algorithm that solves this feasibility problem with a sequence of semidefinite programs whose objectives are designed to encourage low-rank minimizers. Our problem formulation elegantly unifies the configuration space and workspace constraints of a robot: intrinsic robot geometry and obstacle avoidance are both expressed as simple linear matrix equations and inequalities. Our experimental results for a variety of popular manipulator models demonstrate faster and more accurate convergence than a conventional nonlinear optimization-based approach, especially in environments with many obstacles.

Filip Marić, Matthew Giamou, Adam W. Hall et al.

Solving the inverse kinematics problem is a fundamental challenge in motion planning, control, and calibration for articulated robots. Kinematic models for these robots are typically parametrized by joint angles, generating a complicated mapping between the robot configuration and the end-effector pose. Alternatively, the kinematic model and task constraints can be represented using invariant distances between points attached to the robot. In this paper, we formalize the equivalence of distance-based inverse kinematics and the distance geometry problem for a large class of articulated robots and task constraints. Unlike previous approaches, we use the connection between distance geometry and low-rank matrix completion to find inverse kinematics solutions by completing a partial Euclidean distance matrix through local optimization. Furthermore, we parametrize the space of Euclidean distance matrices with the Riemannian manifold of fixed-rank Gram matrices, allowing us to leverage a variety of mature Riemannian optimization methods. Finally, we show that bound smoothing can be used to generate informed initializations without significant computational overhead, improving convergence. We demonstrate that our inverse kinematics solver achieves higher success rates than traditional techniques, and substantially outperforms them on problems that involve many workspace constraints.

Brandon Wagstaff, Valentin Peretroukhin, Jonathan Kelly

Structure from motion (SfM) has recently been formulated as a self-supervised learning problem, where neural network models of depth and egomotion are learned jointly through view synthesis. Herein, we address the open problem of how to best couple, or link, the depth and egomotion network components, so that information such as a common scale factor can be shared between the networks. Towards this end, we introduce several notions of coupling, categorize existing approaches, and present a novel tightly-coupled approach that leverages the interdependence of depth and egomotion at training time and at test time. Our approach uses iterative view synthesis to recursively update the egomotion network input, permitting contextual information to be passed between the components. We demonstrate through substantial experiments that our approach promotes consistency between the depth and egomotion predictions at test time, improves generalization, and leads to state-of-the-art accuracy on indoor and outdoor depth and egomotion evaluation benchmarks.

Jonathan Kelly, Christopher Grebe, Matthew Giamou

We examine the problem of time delay estimation, or temporal calibration, in the context of multisensor data fusion. Differences in processing intervals and other factors typically lead to a relative delay between measurement updates from disparate sensors. Correct (optimal) data fusion demands that the relative delay must either be known in advance or identified online. There have been several recent proposals in the literature to determine the delay using recursive, causal filters such as the extended Kalman filter (EKF). We carefully review this formulation and show that there are fundamental issues with the structure of the EKF (and related algorithms) when the delay is included in the filter state vector as a parameter to be estimated. These structural issues, in turn, leave recursive filters prone to bias and inconsistency. Our theoretical analysis is supported by simulation studies that demonstrate the implications in terms of filter performance; although tuning of the filter noise variances may reduce the chance of inconsistency or divergence, the underlying structural concerns remain. We offer brief suggestions for ways to maintain the computational efficiency of recursive filtering for temporal calibration while avoiding the drawbacks of the standard filtering algorithms.

Trevor Ablett, Yifan Zhai, Jonathan Kelly

Learned visuomotor policies have shown considerable success as an alternative to traditional, hand-crafted frameworks for robotic manipulation. Surprisingly, an extension of these methods to the multiview domain is relatively unexplored. A successful multiview policy could be deployed on a mobile manipulation platform, allowing the robot to complete a task regardless of its view of the scene. In this work, we demonstrate that a multiview policy can be found through imitation learning by collecting data from a variety of viewpoints. We illustrate the general applicability of the method by learning to complete several challenging multi-stage and contact-rich tasks, from numerous viewpoints, both in a simulated environment and on a real mobile manipulation platform. Furthermore, we analyze our policies to determine the benefits of learning from multiview data compared to learning with data collected from a fixed perspective. We show that learning from multiview data results in little, if any, penalty to performance for a fixed-view task compared to learning with an equivalent amount of fixed-view data. Finally, we examine the visual features learned by the multiview and fixed-view policies. Our results indicate that multiview policies implicitly learn to identify spatially correlated features.

Abhinav Grover, Christopher Grebe, Philippe Nadeau et al.

Despite the utility of tactile information, tactile sensors have yet to be widely deployed in industrial robotics settings. Part of the challenge lies in identifying slip and other key events from the tactile data stream. In this paper, we present a learning-based method to detect slip using barometric tactile sensors. Although these sensors have a low resolution, they have many other desirable properties including high reliability and durability, a very slim profile, and a low cost. We are able to achieve slip detection accuracies of greater than 91% while being robust to the speed and direction of the slip motion. Further, we test our detector on two robot manipulation tasks involving common household objects and demonstrate successful generalization to real-world scenarios not seen during training. We show that barometric tactile sensing technology, combined with data-driven learning, is potentially suitable for complex manipulation tasks such as slip compensation.

Emmett Wise, Juraj Peršić, Christopher Grebe et al.

Reliable operation in inclement weather is essential to the deployment of safe autonomous vehicles (AVs). Robustness and reliability can be achieved by fusing data from the standard AV sensor suite (i.e., lidars, cameras) with weather robust sensors, such as millimetre-wavelength radar. Critically, accurate sensor data fusion requires knowledge of the rigid-body transform between sensor pairs, which can be determined through the process of extrinsic calibration. A number of extrinsic calibration algorithms have been designed for 2D (planar) radar sensors - however, recently-developed, low-cost 3D millimetre-wavelength radars are set to displace their 2D counterparts in many applications. In this paper, we present a continuous-time 3D radar-to-camera extrinsic calibration algorithm that utilizes radar velocity measurements and, unlike the majority of existing techniques, does not require specialized radar retroreflectors to be present in the environment. We derive the observability properties of our formulation and demonstrate the efficacy of our algorithm through synthetic and real-world experiments.

Filip Marić, Luka Petrović, Marko Guberina et al.

Articulated robots such as manipulators increasingly must operate in uncertain and dynamic environments where interaction (with human coworkers, for example) is necessary. In these situations, the capacity to quickly adapt to unexpected changes in operational space constraints is essential. At certain points in a manipulator's configuration space, termed singularities, the robot loses one or more degrees of freedom (DoF) and is unable to move in specific operational space directions. The inability to move in arbitrary directions in operational space compromises adaptivity and, potentially, safety. We introduce a geometry-aware singularity index, defined using a Riemannian metric on the manifold of symmetric positive definite matrices, to provide a measure of proximity to singular configurations. We demonstrate that our index avoids some of the failure modes and difficulties inherent to other common indices. Further, we show that this index can be differentiated easily, making it compatible with local optimization approaches used for operational space control. Our experimental results establish that, for reaching and path following tasks, optimization based on our index outperforms a common manipulability maximization technique and ensures singularity-robust motions.

Justin Tomasi, Brandon Wagstaff, Steven L. Waslander et al.

Successful visual navigation depends upon capturing images that contain sufficient useful information. In this letter, we explore a data-driven approach to account for environmental lighting changes, improving the quality of images for use in visual odometry (VO) or visual simultaneous localization and mapping (SLAM). We train a deep convolutional neural network model to predictively adjust camera gain and exposure time parameters such that consecutive images contain a maximal number of matchable features. The training process is fully self-supervised: our training signal is derived from an underlying VO or SLAM pipeline and, as a result, the model is optimized to perform well with that specific pipeline. We demonstrate through extensive real-world experiments that our network can anticipate and compensate for dramatic lighting changes (e.g., transitions into and out of road tunnels), maintaining a substantially higher number of inlier feature matches than competing camera parameter control algorithms.

Filip Marić, Matthew Giamou, Ivan Petrović et al.

The majority of inverse kinematics (IK) algorithms search for solutions in a configuration space defined by joint angles. However, the kinematics of many robots can also be described in terms of distances between rigidly-attached points, which collectively form a Euclidean distance matrix. This alternative geometric description of the kinematics reveals an elegant equivalence between IK and the problem of low-rank matrix completion. We use this connection to implement a novel Riemannian optimization-based solution to IK for various articulated robots with symmetric joint angle constraints.

Alexis Morris, Hallie Siegel, Jonathan Kelly

Building trustworthy autonomous systems is challenging for many reasons beyond simply trying to engineer agents that 'always do the right thing.' There is a broader context that is often not considered within AI and HRI: that the problem of trustworthiness is inherently socio-technical and ultimately involves a broad set of complex human factors and multidimensional relationships that can arise between agents, humans, organizations, and even governments and legal institutions, each with their own understanding and definitions of trust. This complexity presents a significant barrier to the development of trustworthy AI and HRI systems---while systems developers may desire to have their systems 'always do the right thing,' they generally lack the practical tools and expertise in law, regulation, policy and ethics to ensure this outcome. In this paper, we emphasize the "fuzzy" socio-technical aspects of trustworthiness and the need for their careful consideration during both design and deployment. We hope to contribute to the discussion of trustworthy engineering in AI and HRI by i) describing the policy landscape that must be considered when addressing trustworthy computing and the need for usable trust models, ii) highlighting an opportunity for trustworthy-by-design intervention within the systems engineering process, and iii) introducing the concept of a "policy-as-a-service" (PaaS) framework that can be readily applied by AI systems engineers to address the fuzzy problem of trust during the development and (eventually) runtime process. We envision that the PaaS approach, which offloads the development of policy design parameters and maintenance of policy standards to policy experts, will enable runtime trust capabilities intelligent systems in the wild.

Erik Hemberg, Jonathan Kelly, Michal Shlapentokh-Rothman et al.

Many public sources of cyber threat and vulnerability information exist to help defend cyber systems. This paper links MITRE's ATT&CK MATRIX of Tactics and Techniques, NIST's Common Weakness Enumerations (CWE), Common Vulnerabilities and Exposures (CVE), and Common Attack Pattern Enumeration and Classification list (CAPEC), to gain further insight from alerts, threats and vulnerabilities. We preserve all entries and relations of the sources, while enabling bi-directional, relational path tracing within an aggregate data graph called BRON. In one example, we use BRON to enhance the information derived from a list of the top 10 most frequently exploited CVEs. We identify attack patterns, tactics, and techniques that exploit these CVEs and also uncover a disparity in how much linked information exists for each of these CVEs. This prompts us to further inventory BRON's collection of sources to provide a view of the extent and range of the coverage and blind spots of public data sources.

Brandon Wagstaff, Jonathan Kelly

The self-supervised loss formulation for jointly training depth and egomotion neural networks with monocular images is well studied and has demonstrated state-of-the-art accuracy. One of the main limitations of this approach, however, is that the depth and egomotion estimates are only determined up to an unknown scale. In this paper, we present a novel scale recovery loss that enforces consistency between a known camera height and the estimated camera height, generating metric (scaled) depth and egomotion predictions. We show that our proposed method is competitive with other scale recovery techniques that require more information. Further, we demonstrate that our method facilitates network retraining within new environments, whereas other scale-resolving approaches are incapable of doing so. Notably, our egomotion network is able to produce more accurate estimates than a similar method which recovers scale at test time only.