NeuFlow v2: High-Efficiency Optical Flow Estimation on Edge Devices

Real-time high-accuracy optical flow estimation is crucial for various real-world applications. While recent learning-based optical flow methods have achieved high accuracy, they often come with significant computational costs. In this paper, we propose a highly efficient optical flow method that balances high accuracy with reduced computational demands. Building upon NeuFlow v1, we introduce new components including a much more light-weight backbone and a fast refinement module. Both these modules help in keeping the computational demands light while providing close to state of the art accuracy. Compares to other state of the art methods, our model achieves a 10x-70x speedup while maintaining comparable performance on both synthetic and real-world data. It is capable of running at over 20 FPS on 512x384 resolution images on a Jetson Orin Nano. The full training and evaluation code is available at https://github.com/neufieldrobotics/NeuFlow_v2.

Towards Long Term SLAM on Thermal Imagery

Visual SLAM with thermal imagery, and other low contrast visually degraded environments such as underwater, or in areas dominated by snow and ice, remain a difficult problem for many state of the art (SOTA) algorithms. In addition to challenging front-end data association, thermal imagery presents an additional difficulty for long term relocalization and map reuse. The relative temperatures of objects in thermal imagery change dramatically from day to night. Feature descriptors typically used for relocalization in SLAM are unable to maintain consistency over these diurnal changes. We show that learned feature descriptors can be used within existing Bag of Word based localization schemes to dramatically improve place recognition across large temporal gaps in thermal imagery. In order to demonstrate the effectiveness of our trained vocabulary, we have developed a baseline SLAM system, integrating learned features and matching into a classical SLAM algorithm. Our system demonstrates good local tracking on challenging thermal imagery, and relocalization that overcomes dramatic day to night thermal appearance changes. Our code and datasets are available here: https://github.com/neufieldrobotics/IRSLAM_Baseline

On Designing Consistent Covariance Recovery from a Deep Learning Visual Odometry Engine

Deep learning techniques have significantly advanced in providing accurate visual odometry solutions by leveraging large datasets. However, generating uncertainty estimates for these methods remains a challenge. Traditional sensor fusion approaches in a Bayesian framework are well-established, but deep learning techniques with millions of parameters lack efficient methods for uncertainty estimation. This paper addresses the issue of uncertainty estimation for pre-trained deep-learning models in monocular visual odometry. We propose formulating a factor graph on an implicit layer of the deep learning network to recover relative covariance estimates, which allows us to determine the covariance of the Visual Odometry (VO) solution. We showcase the consistency of the deep learning engine's covariance approximation with an empirical analysis of the covariance model on the EUROC datasets to demonstrate the correctness of our formulation.

NeuFlow: Real-time, High-accuracy Optical Flow Estimation on Robots Using Edge Devices

Real-time high-accuracy optical flow estimation is a crucial component in various applications, including localization and mapping in robotics, object tracking, and activity recognition in computer vision. While recent learning-based optical flow methods have achieved high accuracy, they often come with heavy computation costs. In this paper, we propose a highly efficient optical flow architecture, called NeuFlow, that addresses both high accuracy and computational cost concerns. The architecture follows a global-to-local scheme. Given the features of the input images extracted at different spatial resolutions, global matching is employed to estimate an initial optical flow on the 1/16 resolution, capturing large displacement, which is then refined on the 1/8 resolution with lightweight CNN layers for better accuracy. We evaluate our approach on Jetson Orin Nano and RTX 2080 to demonstrate efficiency improvements across different computing platforms. We achieve a notable 10x-80x speedup compared to several state-of-the-art methods, while maintaining comparable accuracy. Our approach achieves around 30 FPS on edge computing platforms, which represents a significant breakthrough in deploying complex computer vision tasks such as SLAM on small robots like drones. The full training and evaluation code is available at https://github.com/neufieldrobotics/NeuFlow.

OASIS: Optimal Arrangements for Sensing in SLAM

The number and arrangement of sensors on an autonomous mobile robot dramatically influence its perception capabilities. Ensuring that sensors are mounted in a manner that enables accurate detection, localization, and mapping is essential for the success of downstream control tasks. However, when designing a new robotic platform, researchers and practitioners alike usually mimic standard configurations or maximize simple heuristics like field-of-view (FOV) coverage to decide where to place exteroceptive sensors. In this work, we conduct an information-theoretic investigation of this overlooked element of mobile robotic perception in the context of simultaneous localization and mapping (SLAM). We show how to formalize the sensor arrangement problem as a form of subset selection under the E-optimality performance criterion. While this formulation is NP-hard in general, we further show that a combination of greedy sensor selection and fast convex relaxation-based post-hoc verification enables the efficient recovery of certifiably optimal sensor designs in practice. Results from synthetic experiments reveal that sensors placed with OASIS outperform benchmarks in terms of mean squared error of visual SLAM estimates.

Direct Superpoints Matching for Fast and Robust Point Cloud Registration

Although deep neural networks endow the downsampled superpoints with discriminative feature representations, directly matching them is usually not used alone in state-of-the-art methods, mainly for two reasons. First, the correspondences are inevitably noisy, so RANSAC-like refinement is usually adopted. Such ad hoc postprocessing, however, is slow and not differentiable, which can not be jointly optimized with feature learning. Second, superpoints are sparse and thus more RANSAC iterations are needed. Existing approaches use the coarse-to-fine strategy to propagate the superpoints correspondences to the point level, which are not discriminative enough and further necessitates the postprocessing refinement. In this paper, we present a simple yet effective approach to extract correspondences by directly matching superpoints using a global softmax layer in an end-to-end manner, which are used to determine the rigid transformation between the source and target point cloud. Compared with methods that directly predict corresponding points, by leveraging the rich information from the superpoints matchings, we can obtain more accurate estimation of the transformation and effectively filter out outliers without any postprocessing refinement. As a result, our approach is not only fast, but also achieves state-of-the-art results on the challenging ModelNet and 3DMatch benchmarks. Our code and model weights will be publicly released.

Challenges of Indoor SLAM: A multi-modal multi-floor dataset for SLAM evaluation

Robustness in Simultaneous Localization and Mapping (SLAM) remains one of the key challenges for the real-world deployment of autonomous systems. SLAM research has seen significant progress in the last two and a half decades, yet many state-of-the-art (SOTA) algorithms still struggle to perform reliably in real-world environments. There is a general consensus in the research community that we need challenging real-world scenarios which bring out different failure modes in sensing modalities. In this paper, we present a novel multi-modal indoor SLAM dataset covering challenging common scenarios that a robot will encounter and should be robust to. Our data was collected with a mobile robotics platform across multiple floors at Northeastern University's ISEC building. Such a multi-floor sequence is typical of commercial office spaces characterized by symmetry across floors and, thus, is prone to perceptual aliasing due to similar floor layouts. The sensor suite comprises seven global shutter cameras, a high-grade MEMS inertial measurement unit (IMU), a ZED stereo camera, and a 128-channel high-resolution lidar. Along with the dataset, we benchmark several SLAM algorithms and highlight the problems faced during the runs, such as perceptual aliasing, visual degradation, and trajectory drift. The benchmarking results indicate that parts of the dataset work well with some algorithms, while other data sections are challenging for even the best SOTA algorithms. The dataset is available at https://github.com/neufieldrobotics/NUFR-M3F.

Bridging the Domain Gap between Synthetic and Real-World Data for Autonomous Driving

Modern autonomous systems require extensive testing to ensure reliability and build trust in ground vehicles. However, testing these systems in the real-world is challenging due to the lack of large and diverse datasets, especially in edge cases. Therefore, simulations are necessary for their development and evaluation. However, existing open-source simulators often exhibit a significant gap between synthetic and real-world domains, leading to deteriorated mobility performance and reduced platform reliability when using simulation data. To address this issue, our Scoping Autonomous Vehicle Simulation (SAVeS) platform benchmarks the performance of simulated environments for autonomous ground vehicle testing between synthetic and real-world domains. Our platform aims to quantify the domain gap and enable researchers to develop and test autonomous systems in a controlled environment. Additionally, we propose using domain adaptation technologies to address the domain gap between synthetic and real-world data with our SAVeS$^+$ extension. Our results demonstrate that SAVeS$^+$ is effective in helping to close the gap between synthetic and real-world domains and yields comparable performance for models trained with processed synthetic datasets to those trained on real-world datasets of same scale. This paper highlights our efforts to quantify and address the domain gap between synthetic and real-world data for autonomy simulation. By enabling researchers to develop and test autonomous systems in a controlled environment, we hope to bring autonomy simulation one step closer to realization.

Bayesian Decision Making to Localize Visual Queries in 2D

This report describes our approach for the EGO4D 2023 Visual Query 2D Localization Challenge. Our method aims to reduce the number of False Positives (FP) that occur because of high similarity between the visual crop and the proposed bounding boxes from the baseline's Region Proposal Network (RPN). Our method uses a transformer to determine similarity in higher dimensions which is used as our prior belief. The results are then combined together with the similarity in lower dimensions from the Siamese Head, acting as our measurement, to generate a posterior which is then used to determine the final similarity of the visual crop with the proposed bounding box. Our code is publicly available $\href{https://github.com/s-m-asjad/EGO4D_VQ2D}{here}$.

Design and Evaluation of a Generic Visual SLAM Framework for Multi-Camera Systems

Multi-camera systems have been shown to improve the accuracy and robustness of SLAM estimates, yet state-of-the-art SLAM systems predominantly support monocular or stereo setups. This paper presents a generic sparse visual SLAM framework capable of running on any number of cameras and in any arrangement. Our SLAM system uses the generalized camera model, which allows us to represent an arbitrary multi-camera system as a single imaging device. Additionally, it takes advantage of the overlapping fields of view (FoV) by extracting cross-matched features across cameras in the rig. This limits the linear rise in the number of features with the number of cameras and keeps the computational load in check while enabling an accurate representation of the scene. We evaluate our method in terms of accuracy, robustness, and run time on indoor and outdoor datasets that include challenging real-world scenarios such as narrow corridors, featureless spaces, and dynamic objects. We show that our system can adapt to different camera configurations and allows real-time execution for typical robotic applications. Finally, we benchmark the impact of the critical design parameters - the number of cameras and the overlap between their FoV that define the camera configuration for SLAM. All our software and datasets are freely available for further research.