QuadricSLAM: Dual Quadrics From Object Detections as Landmarks in Object-Oriented SLAM

Nicholson, Lachlan; Milford, Michael; Sünderhauf, Niko

doi:10.1109/lra.2018.2866205

Cited by 238 publications

(250 citation statements)

References 29 publications

Supporting

Mentioning

248

Contrasting

Order By: Relevance

“…The case of perspective projection is considered in [22]. However, as we mentioned in the introduction, the SLAM resolution is constrained by odometry measurements.…”

Section: More Related Workmentioning

confidence: 99%

“…The first term reflects the attachment to odometry measurements and the second term is a geometric error between the projected ellipsoids and the detection BBs (see [22] for the details of the notations). The geometric error is defined as the sum of squares of the distances between borders of the BBs of the reprojected ellipsoids and borders of the detection BBs.…”

Section: More Related Workmentioning

confidence: 99%

“…tional computer vision problems based on high-level primitives (objects), instead of the traditional low-level primitives [16,17,22,28]. However, it is not easy to accurately compute the camera pose from coarse 2D BBs of objects, particularly when few objects are detected.…”

mentioning

confidence: 99%

“…The authors show that a closed-form solution exists from three views. In [22], Nicholson et al describe a complete SLAM (Simultaneous Localization And Mapping) system that is able to reconstruct ellipsoids in 3D and simultaneously compute the camera poses from several ellipses fitted, again, to the object detection BBs. YOLOv3 [26] is used for object detection.…”

mentioning

confidence: 99%

“…In this paper, we build on the advances made by Rubino et al [28] and Nicholson et al [22] to design a robust, portable and potentially sensor-less method that estimates the pose of a camera at object level (see Fig. 1).…”

mentioning

confidence: 99%

See 4 more Smart Citations

Camera Relocalization with Ellipsoidal Abstraction of Objects

Gaudillière

Simon

Berger

2019

2019 IEEE International Symposium on Mixed and Augmented Reality (ISMAR)

View full text Add to dashboard Cite

“…The case of perspective projection is considered in [22]. However, as we mentioned in the introduction, the SLAM resolution is constrained by odometry measurements.…”

Section: More Related Workmentioning

confidence: 99%

Section: More Related Workmentioning

confidence: 99%

mentioning

confidence: 99%

mentioning

confidence: 99%

mentioning

confidence: 99%

See 3 more Smart Citations

Camera Relocalization with Ellipsoidal Abstraction of Objects

Gaudillière

Simon

Berger

2019

2019 IEEE International Symposium on Mixed and Augmented Reality (ISMAR)

View full text Add to dashboard Cite

Textile Technology for Soft Robotic and Autonomous Garments

Sánchez¹,

Walsh²,

Wood³

2020

Adv Funct Materials

168

123

View full text Add to dashboard Cite

Textiles have emerged as a promising class of materials for developing wearable robots that move and feel like everyday clothing. Textiles represent a favorable material platform for wearable robots due to their flexibility, low weight, breathability, and soft hand‐feel. Textiles also offer a unique level of programmability because of their inherent hierarchical nature, enabling researchers to modify and tune properties at several interdependent material scales. With these advantages and capabilities in mind, roboticists have begun to use textiles, not simply as substrates, but as functional components that program actuation and sensing. In parallel, materials scientists are developing new materials that respond to thermal, electrical, and hygroscopic stimuli by leveraging textile structures for function. Although textiles are one of humankind's oldest technologies, materials scientists and roboticists are just beginning to tap into their potential. This review provides a textile‐centric survey of the current state of the art in wearable robotic garments and highlights metrics that will guide materials development. Recent advances in textile materials for robotic components (i.e., as sensors, actuators, and integration components) are described with a focus on how these materials and technologies set the stage for wearable robots programmed at the material level.

show abstract

ATG-PVD: Ticketing Parking Violations on a Drone

Wang

Liu

Huang

et al. 2020

Lecture Notes in Computer Science

View full text Add to dashboard Cite

In this paper, we introduce a novel suspect-and-investigate framework, which can be easily embedded in a drone for automated parking violation detection (PVD). Our proposed framework consists of: 1) SwiftFlow, an efficient and accurate convolutional neural network (CNN) for unsupervised optical flow estimation; 2) Flow-RCNN, a flow-guided CNN for car detection and classification; and 3) an illegally parked car (IPC) candidate investigation module developed based on visual SLAM. The proposed framework was successfully embedded in a drone from ATG Robotics. The experimental results demonstrate that, firstly, our proposed SwiftFlow outperforms all other state-of-the-art unsupervised optical flow estimation approaches in terms of both speed and accuracy; secondly, IPC candidates can be effectively and efficiently detected by our proposed Flow-RCNN, with a better performance than our baseline network, Faster-RCNN; finally, the actual IPCs can be successfully verified by our investigation module after drone re-localization.

show abstract

QuadricSLAM: Dual Quadrics From Object Detections as Landmarks in Object-Oriented SLAM

Cited by 238 publications

References 29 publications

Camera Relocalization with Ellipsoidal Abstraction of Objects

Camera Relocalization with Ellipsoidal Abstraction of Objects

Textile Technology for Soft Robotic and Autonomous Garments

ATG-PVD: Ticketing Parking Violations on a Drone

Contact Info

Product

Resources

About