Device Pose Research Articles

An automatic image-to-DEM alignment approach for annotating mountains pictures on a smartphone

Augmented reality (AR) technologies enhance users knowledge of their immediate surroundings by presenting contextualized and spatially relevant information. This augmentation is enabled by the automatic estimation of the device's pose with respect to the environment. Many simple AR applications for smartphones are well served by the pose estimate provided by GPS and inertial sensors, while other demand an higher accuracy which can be obtained with vision-based methods. The application presented in this paper, which aims to augment pictures of mountainous landscapes with geo-referenced data, belongs to the latter category. Our application is based on a novel approach for image-to-world registration which jointly relies on inertial and visual sensors. In a nutshell, first GPS and inertial sensors are used to compute a rough estimate of the device position and pose, then visual data are employed to refine it. Specifically, a learning-based edge detection algorithm is used to extract mountain profiles from the picture of interest. Then, a novel registration algorithm based on a robust optimization framework is employed to align the computed profiles to synthetic ones obtained from Digital Elevation Models. Our experiments, conducted on a novel dataset of manually aligned pictures which is made publicly available, demonstrate that the proposed registration method guarantees an increased accuracy with respect to competing methods and it is computationally efficient when implemented on a smartphone.

Tangible and modular input device for character articulation

Articulation of 3D characters requires control over many degrees of freedom: a difficult task with standard 2D interfaces. We present a tangible input device composed of interchangeable, hot-pluggable parts. Embedded sensors measure the device's pose at rates suitable for real-time editing and animation. Splitter parts allow branching to accommodate any skeletal tree. During assembly, the device recognizes topological changes as individual parts or pre-assembled subtrees are plugged and unplugged. A novel semi-automatic registration approach helps the user quickly map the device's degrees of freedom to a virtual skeleton inside the character. User studies report favorable comparisons to mouse and keyboard interfaces for the tasks of target acquisition and pose replication. Our device provides input for character rigging and automatic weight computation, direct skeletal deformation, interaction with physical simulations, and handle-based variational geometric modeling.

A new interface using image-based foot tracking for motion sensing devices

Foot interfacing has applications in games that require foot manipulation control using motion sensing devices. However, when part of a foot overlaps itself (selfocclusion), is out of the line of sight of the infrared sensor and it is then difficult to accurately capture foot pose. To overcome this problem, we propose a real-time motion sensing method based on 3-dimensional motion estimation. Our method employs pose parameters that are acquired using the ground plane and camera coordinates. This enables an assessment of the positions of the two joints that comprise a skeletal foot model. The proposed approach can capture a foot's 3-dimensional information that an infrared sensor may miss; therefore, motion sensing device's pose estimation results improve. According to the tracking and experimental results, the proposed method is applicable for use by motion sensing devices.

Manual-scanning optical coherence tomography probe based on position tracking

A method based on position tracking to reconstruct images for a manual-scanning optical coherence tomography (OCT) probe is proposed and implemented. The method employs several feature points on a hand-held probe and a camera to track the device's pose. The continuous device poses tracking, and the collected OCT depth scans can then be combined to render OCT images. The tracking accuracy of the system was characterized to be about 6 microm along two axes and 19 microm along the third. A phantom target was used to validate the method. In addition, we report OCT images of a 54-stage Xenopus laevis tadpole acquired by manual scanning.