VersaVIS—An Open Versatile Multi-Camera Visual-Inertial Sensor Suite

Tschopp, Florian; Riner, Michael; Bernreiter, Lukas; Furrer, Fadri; Novković, Tonči; Pfrunder, Andreas; Cadena, César; Siegwart, Roland; Nieto, Juan

doi:10.3390/s20051439

Cited by 34 publications

(22 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This timing issue hindered the ML implementation when combining separate sensors and limited a valid computation of temporal and spatial gait features. Some researchers have worked to solve this problem through the use of highly customized wirelessly networked IMUs or through the use of extensive camera systems connected to real-time references to guarantee accurate kinematics data collection for different purposes [ 39 , 40 ]. These studies do provide more reliable results than our study, but at a much greater cost of time and resources, and restrictions in deploying the systems to the general clinical setting.…”

Section: Discussionmentioning

confidence: 99%

Machine Learning Prediction of Fall Risk in Older Adults Using Timed Up and Go Test Kinematics

Roshdibenam

Jogerst

Butler

et al. 2021

Sensors

View full text Add to dashboard Cite

Falls among the elderly population cause detrimental physical, mental, financial problems and, in the worst case, death. The increasing number of people entering the higher risk age-range has increased clinicians’ attention to intervene. Clinical tools, e.g., the Timed Up and Go (TUG) test, have been created for aiding clinicians in fall-risk assessment. Often simple to evaluate, these assessments are subject to a clinician’s judgment. Wearable sensor data with machine learning algorithms were introduced as an alternative to precisely quantify ambulatory kinematics and predict prospective falls. However, they require a long-term evaluation of large samples of subjects’ locomotion and complex feature engineering of sensor kinematics. Therefore, it is critical to build an objective fall-risk detection model that can efficiently measure biometric risk factors with minimal costs. We built and studied a sensor data-driven convolutional neural network model to predict older adults’ fall-risk status with relatively high sensitivity to geriatrician’s expert assessment. The sample in this study is representative of older patients with multiple co-morbidity seen in daily medical practice. Three non-intrusive wearable sensors were used to measure participants’ gait kinematics during the TUG test. This data collection ensured convenient capture of various gait impairment aspects at different body locations.

show abstract

Section: Discussionmentioning

confidence: 99%

Machine Learning Prediction of Fall Risk in Older Adults Using Timed Up and Go Test Kinematics

Roshdibenam

Jogerst

Butler

et al. 2021

Sensors

View full text Add to dashboard Cite

show abstract

“…On our platform, these are not hardware synchronized. Some cameras can be tightly synchronized, while triggering boards such as the VersaVIS board exist [30] which are able to synchronize several cameras and IMUs. Extending these to cover other types of sensors, such as joint encoders, would improve the usability and accuracy of our proposed method.…”

Section: Discussionmentioning

confidence: 99%

Semi-automatic 3D Object Keypoint Annotation and Detection for the Masses

Blomqvist¹,

Chung²,

Ott³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Creating computer vision datasets requires careful planning and lots of time and effort. In robotics research, we often have to use standardized objects, such as the YCB object set, for tasks such as object tracking, pose estimation, grasping and manipulation, as there are datasets and pre-learned methods available for these objects. This limits the impact of our research since learning-based computer vision methods can only be used in scenarios that are supported by existing datasets. In this work, we present a full object keypoint tracking toolkit, encompassing the entire process from data collection, labeling, model learning and evaluation. We present a semi-automatic way of collecting and labeling datasets using a wrist mounted camera on a standard robotic arm. Using our toolkit and method, we are able to obtain a working 3D object keypoint detector and go through the whole process of data collection, annotation and learning in just a couple hours of active time.

show abstract

“…The Arduino is now capable of collecting information from various sensors and sending real-time commands to the motors for precise control. Furthermore, by also running VersaVIS [177] on this microcontroller, the current gimbal positions can be precisely timestamped. High-level commands can still be sent from the computer via ROS and ROSserial [178] to the Arduino, which are processed and sent on to the motors.…”

Section: Control Systemsmentioning

confidence: 99%

High-Precision Low-Cost Gimballing Platform for Long-Range Railway Obstacle Detection

Assaf

Cornelius

Cadena

et al. 2022

Sensors

Self Cite

View full text Add to dashboard Cite

Increasing demand for rail transportation results transportation by rail, resulting in denser and more high-speed usage of the existing railway network, making makes new and more advanced vehicle safety systems necessary. Furthermore, high traveling speeds and the greatlarge weights of trains lead to long braking distances—all of which necessitates Long braking distances, due to high travelling speeds and the massive weight of trains, necessitate a Long-Range Obstacle Detection (LROD) system, capable of detecting humans and other objects more than 1000 m in advance. According to current research, only a few sensor modalities are capable of reaching this far and recording sufficiently accurate enoughdata to distinguish individual objects. The limitation of these sensors, such as a 1D-Light Detection and Ranging (LiDAR), is however a very narrow Field of View (FoV), making it necessary to use ahigh-precision means of orienting to target them at possible areas of interest. To close this research gap, this paper presents a novel approach to detecting railway obstacles by developinga high-precision pointing mechanism, for the use in a future novel railway obstacle detection system In this work such a high-precision pointing mechanism is developed, capable of targeting aiming a 1D-LiDAR at humans or objects at the required distance. This approach addresses To address the challenges of a low target pricelimited budget, restricted access to high-precision machinery and equipment as well as unique requirements of our target application., a novel pointing mechanism has been designed and developed. By combining established elements from 3D printers and Computer Numerical Control (CNC) machines with a double-hinged lever system, simple and cheaplow-cost components are capable of precisely orienting an arbitrary sensor platform. The system’s actual pointing accuracy has been evaluated using a controlled, in-door, long-range experiment. The device was able to demonstrate a precision of 6.179 mdeg, which is at the limit of the measurable precision of the designed experiment.

show abstract

VersaVIS—An Open Versatile Multi-Camera Visual-Inertial Sensor Suite

Cited by 34 publications

References 20 publications

Machine Learning Prediction of Fall Risk in Older Adults Using Timed Up and Go Test Kinematics

Machine Learning Prediction of Fall Risk in Older Adults Using Timed Up and Go Test Kinematics

Semi-automatic 3D Object Keypoint Annotation and Detection for the Masses

High-Precision Low-Cost Gimballing Platform for Long-Range Railway Obstacle Detection

Contact Info

Product

Resources

About