scholarly journals Unsupervised Learning of Depth and Ego-Motion from Cylindrical Panoramic Video with Applications for Virtual Reality

2020 ◽  
Vol 14 (03) ◽  
pp. 333-356
Author(s):  
Alisha Sharma ◽  
Ryan Nett ◽  
Jonathan Ventura

We introduce a convolutional neural network model for unsupervised learning of depth and ego-motion from cylindrical panoramic video. Panoramic depth estimation is an important technology for applications such as virtual reality, 3D modeling, and autonomous robotic navigation. In contrast to previous approaches for applying convolutional neural networks to panoramic imagery, we use the cylindrical panoramic projection which allows for the use of the traditional CNN layers such as convolutional filters and max pooling without modification. Our evaluation of synthetic and real data shows that unsupervised learning of depth and ego-motion on cylindrical panoramic images can produce high-quality depth maps and that an increased field-of-view improves ego-motion estimation accuracy. We create two new datasets to evaluate our approach: a synthetic dataset created using the CARLA simulator, and Headcam, a novel dataset of panoramic video collected from a helmet-mounted camera while biking in an urban setting. We also apply our network to the problem of converting monocular panoramas to stereo panoramas.

2019 ◽  
Vol 39 (2) ◽  
pp. 543-570 ◽  
Author(s):  
Mingyang Geng ◽  
Suning Shang ◽  
Bo Ding ◽  
Huaimin Wang ◽  
Pengfei Zhang

2007 ◽  
Vol 10 (4) ◽  
pp. 508-515 ◽  
Author(s):  
Mary F. Macedonio ◽  
Thomas D. Parsons ◽  
Raymond A. Digiuseppe ◽  
Brenda A. Weiderhold ◽  
Albert A. Rizzo

2017 ◽  
Vol 24 (6) ◽  
pp. 1283-1295 ◽  
Author(s):  
Tomáš Faragó ◽  
Petr Mikulík ◽  
Alexey Ershov ◽  
Matthias Vogelgesang ◽  
Daniel Hänschke ◽  
...  

An open-source framework for conducting a broad range of virtual X-ray imaging experiments,syris, is presented. The simulated wavefield created by a source propagates through an arbitrary number of objects until it reaches a detector. The objects in the light path and the source are time-dependent, which enables simulations of dynamic experiments,e.g.four-dimensional time-resolved tomography and laminography. The high-level interface ofsyrisis written in Python and its modularity makes the framework very flexible. The computationally demanding parts behind this interface are implemented in OpenCL, which enables fast calculations on modern graphics processing units. The combination of flexibility and speed opens new possibilities for studying novel imaging methods and systematic search of optimal combinations of measurement conditions and data processing parameters. This can help to increase the success rates and efficiency of valuable synchrotron beam time. To demonstrate the capabilities of the framework, various experiments have been simulated and compared with real data. To show the use case of measurement and data processing parameter optimization based on simulation, a virtual counterpart of a high-speed radiography experiment was created and the simulated data were used to select a suitable motion estimation algorithm; one of its parameters was optimized in order to achieve the best motion estimation accuracy when applied on the real data.syriswas also used to simulate tomographic data sets under various imaging conditions which impact the tomographic reconstruction accuracy, and it is shown how the accuracy may guide the selection of imaging conditions for particular use cases.


Author(s):  
J. Parker Mitchell ◽  
Grant Bruer ◽  
Mark E. Dean ◽  
James S. Plank ◽  
Garrett S. Rose ◽  
...  

Author(s):  
Gregorio Soria ◽  
L. M. Ortega Alvarado ◽  
Francisco R. Feito

Augmented reality (AR) has experienced a breakthrough in many areas of application thanks to cheaper hardware and a strong industry commitment. In the field of management of urban facilities, this technology allows virtual access and interaction with hidden underground elements. This paper presents a new approach to enable AR in mobile devices such as Google Tango, which has specific capabilities to be used outdoors. The first objective is to provide full functionality in the life-cycle management of subsoil infrastructures through this technology. This implies not only visualization, interaction, and free navigation, but also editing, deleting, and inserting elements ubiquitously. For this, a topological data model for three-dimensional (3D) data has been designed. Another important contribution of the paper is getting exact location and orientation performed in only a few minutes, using no additional markers or hardware. This accuracy in the initial positioning, together with the device sensing, avoids the usual errors during the navigation process in AR. Similar functionality has also been implemented in a nonubiquitous way to be supported by any other device through virtual reality (VR). The tests have been performed using real data of the city of Jaén (Spain).


2019 ◽  
Vol 78 ◽  
pp. 284-292 ◽  
Author(s):  
Renyue Dai ◽  
Yongbin Gao ◽  
Zhijun Fang ◽  
Xiaoyan Jiang ◽  
Anjie Wang ◽  
...  

Sign in / Sign up

Export Citation Format

Share Document