Mo Shan

University of California, San Diego | UCSD · Department of Electrical and Computer Engineering

Autonomous systems need to understand the semantics and geometry of their surroundings in order to comprehend and safely execute object-level task specifications. This paper proposes an expressive yet compact model for joint object pose and shape optimization, and an associated optimization algorithm to infer an object-level map from multi-view RGB...

This paper focuses on building semantic maps, containing object poses and shapes, using a monocular camera. This is an important problem because robots need rich understanding of geometry and context if they are to shape the future of transportation, construction, and agriculture. Our contribution is an instance-specific mesh model of object shape...

Introducing object-level semantic information into simultaneous localization and mapping (SLAM) system is critical. It not only improves the performance but also enables tasks specified in terms of meaningful objects. This work presents OrcVIO, for visual-inertial odometry tightly coupled with tracking and optimization over structured object models...

A salient object detection approach is proposed in this paper, tailored to the aerial images collected by Unmanned Aerial Vehicles (UAVs). In particular, the aerial images are classified. The selected image is segmented using superpixel-s, and then a weak saliency map is constructed based on image priors. Positive and negative samples are selected...

With the increasing need for micro aerial vehicles (MAVs) to work in GPS-denied environments, vision technique has been extensively explored to realize robust flight control. In this paper, we propose to employ Google map, which could be outdated due to the lag of updating, as reference to realize robust navigation for MAVs equipped with a downward...

Keypoint detection using convolutional neural networks (CNNs) requires a large amount of annotations that are time consuming and labor intensive. In this work, it is shown that CNNs could merely rely on class labels to categorize images and locate the keypoints simultaneously. Specifically, keypoints are detected in a multiscale framework based on...

High level understanding of sequential visual input is important for safe and stable autonomy, especially in localization and object detection. While traditional object classification and tracking approaches are specifically designed to handle variations in rotation and scale, current state-of-the-art approaches based on deep learning achieve bette...

We propose a framework for Google Map aided UAV navigation in GPS-denied environment. Geo-referenced navigation provides drift-free localization and does not require loop closures. The UAV position is initialized via correlation, which is simple and efficient. We then use optical flow to predict its position in subsequent frames. During pose tracki...

Online localization and mapping in unknown environment is essential for Micro Aerial Vehicles (MAVs). Both accuracy and robustness are required in the realtime applications. In this paper, we present a dual camera system to estimate the pose of MAV and generate an obstacle map for navigation. The recently released lightweight Intel Realsense depth...

Simultaneous Localization and Mapping (SLAM) refers to the problem of using various sensors like laser scanner, RGB cameras, RGB-D cameras, etc, to estimate the position of the robot, and concurrently construct the 2D/3D map of the environment. The SLAM community has made great progress in the past few decades. So far the 2D SLAM problem with range...

Inspired by the outstanding performance of sparse representation (SR) in a variety of image/video relevant classification and identification tasks, we propose an adaptive SR method for painting style analysis. Significantly improved over previous SRbased methods, which heavily rely on the comparison of query paintings, our method is able to authent...

Sparse coding is making significant impact in computer vision and signal processing communities, which achieves the state-of-the-art performance in a variety of applications for images, e.g. denoising, restoration and synthesis, etc. We propose an adaptive and robust sparse coding algorithm exploiting the characteristics of typical laser range data...

Inspired by the outstanding performance of sparse coding in applications of image denoising, restoration, classification, etc., we propose an adaptive sparse coding method for painting style analysis that is traditionally carried out by art connoisseurs and experts. Significantly improved over previous sparse coding methods, which heavily rely on t...

This paper is composed of two parts: firstly to bring home to readers a comprehensive and authoritative introduction of the achievements in the field of video surveillance, specifically, of the software algorithms. We review the state-of-the-art techniques ranging from the preliminary visual processing of foreground detection, to mid-level object t...

Recent evaluation of representative background subtraction techniques demonstrated the drawbacks of these methods, with hardly any approach being able to reach more than 50% precision at recall level higher than 90%. Challenges in realistic environment include illumination change causing complex intensity variation, background motions (trees, waves...