Related papers: MAC-VO: Metrics-aware Covariance for Learning-base…
Recently, there are emerging many stereo matching methods for autonomous driving based on unsupervised learning. Most of them take advantage of reconstruction losses to remove dependency on disparity groundtruth. Occlusion handling is a…
This paper introduces a novel framework called Mode-wise Principal Subspace Pursuit (MOP-UP) to extract hidden variations in both the row and column dimensions for matrix data. To enhance the understanding of the framework, we introduce a…
Visual-inertial odometry (VIO) systems traditionally rely on filtering or optimization-based techniques for egomotion estimation. While these methods are accurate under nominal conditions, they are prone to failure during severe…
Localization is an indispensable component of a robot's autonomy stack that enables it to determine where it is in the environment, essentially making it a precursor for any action execution or planning. Although convolutional neural…
Inspired by the cognitive process of humans and animals, Curriculum Learning (CL) trains a model by gradually increasing the difficulty of the training data. In this paper, we study whether CL can be applied to complex geometry problems…
This paper proposes a multiple-model adaptive control methodology, using set-valued observers (MMAC-SVO) for the identification subsystem, that is able to provide robust stability and performance guarantees for the closed-loop, when the…
Dense visual odometry (VO), which provides pose estimation and dense 3D reconstruction, serves as the cornerstone for applications ranging from robotics to augmented reality. Recently, feed-forward models have demonstrated remarkable…
Visual-Inertial Odometry (VIO) is the problem of estimating a robot's trajectory by combining information from an inertial measurement unit (IMU) and a camera, and is of great interest to the robotics community. This paper develops a novel…
Many interesting tasks in machine learning and computer vision are learned by optimising an objective function defined as a weighted linear combination of multiple losses. The final performance is sensitive to choosing the correct…
Recent work in multi-view stereo (MVS) combines learnable photometric scores and regularization with PatchMatch-based optimization to achieve robust pixelwise estimates of depth, normals, and visibility. However, non-learning based methods…
Controlling the radar beam-pattern by optimizing the transmit covariance matrix is a well-established approach for performance enhancement in multiple-input-multiple-output (MIMO) radars. In this paper, we investigate the joint optimization…
Metric Cross-View Geo-Localization (MCVGL) aims to estimate the 3-DoF camera pose (position and heading) by matching ground and satellite images. In this work, instead of pinhole and satellite images, we study robust MCVGL using holistic…
We propose to leverage the local information in image sequences to support global camera relocalization. In contrast to previous methods that regress global poses from single images, we exploit the spatial-temporal consistency in sequential…
Event-based cameras asynchronously capture individual visual changes in a scene. This makes them more robust than traditional frame-based cameras to highly dynamic motions and poor illumination. It also means that every measurement in a…
Audio-Visual Segmentation (AVS) aims to identify, at the pixel level, the object in a visual scene that produces a given sound. Current AVS methods rely on costly fine-grained annotations of mask-audio pairs, making them impractical for…
PointGoal navigation in indoor environment is a fundamental task for personal robots to navigate to a specified point. Recent studies solved this PointGoal navigation task with near-perfect success rate in photo-realistically simulated…
Visual-inertial odometry (VIO) is widely used in various fields, such as robots, drones, and autonomous vehicles. However, real-world scenes often feature dynamic objects, compromising the accuracy of VIO. The diversity and partial…
Multimodal learning has been lacking principled ways of combining information from different modalities and learning a low-dimensional manifold of meaningful representations. We study multimodal learning and sensor fusion from a latent…
Deep learning approaches for Visual-Inertial Odometry (VIO) have proven successful, but they rarely focus on incorporating robust fusion strategies for dealing with imperfect input sensory data. We propose a novel end-to-end selective…
This paper addresses the challenge of improving learning-based monocular visual odometry (VO) in underwater environments by integrating principles of underwater optical imaging to manipulate optical flow estimation. Leveraging the inherent…