Fengting YangPennsylvania State University | Penn State · College of Information Sciences and Technology
Fengting Yang
PhD Candidate
About
12
Publications
2,799
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
237
Citations
Citations since 2017
Introduction
Skills and Expertise
Additional affiliations
August 2017 - present
September 2014 - February 2016
Publications
Publications (12)
Non-orthogonal shaft laser theodolite (N-theodolite) is a new kind of large-scale metrological instrument made up by two rotary tables and one collimated laser. There are three axes for an N-theodolite. According to naming conventions in traditional theodolite, rotary axes of two rotary tables are called as horizontal axis and vertical axis, respec...
A novel cost-effective non-contact 3D measurement system is proposed in this paper, which consists of two rotary tables and one laser range finder. No orthogonal accuracy between the three axes (two rotation axes and the laser axis) is required, i.e. the three parts of the sensor unit (two rotary tables and the laser range finder) need not be assem...
In computer vision, superpixels have been widely used as an effective way to reduce the number of image primitives for subsequent processing. But only a few attempts have been made to incorporate them into deep neural networks. One main reason is that the standard convolution operation is defined on regular grids and becomes inefficient when applie...
In this paper, we study the problem of recovering 3D planar surfaces from a single image of man-made environment. We show that it is possible to directly train a deep neural network to achieve this goal. A novel plane structure-induced loss is proposed to train the network to simultaneously predict a plane segmentation map and the parameters of the...
We present PlanarRecon -- a novel framework for globally coherent detection and reconstruction of 3D planes from a posed monocular video. Unlike previous works that detect planes in 2D from a single image, PlanarRecon incrementally detects planes in 3D for each video fragment, which consists of a set of key frames, from a volumetric representation...
Depth-from-focus (DFF) is a technique that infers depth using the focus change of a camera. In this work, we propose a convolutional neural network (CNN) to find the best-focused pixels in a focal stack and infer depth from the focus estimation. The key innovation of the network is the novel deep differential focus volume (DFV). By computing the fi...
In computer vision, superpixels have been widely used as an effective way to reduce the number of image primitives for subsequent processing. But only a few attempts have been made to incorporate them into deep neural networks. One main reason is that the standard convolution operation is defined on regular grids and becomes inefficient when applie...
In this paper, we study the problem of recovering 3D planar surfaces from a single image of man-made environment. We show that it is possible to directly train a deep neural network to achieve this goal. A novel plane structure-induced loss is proposed to train the network to simultaneously predict a plane segmentation map and the parameters of the...
According to the Non-orthogonal total station system principle, the main sources of the system measurement error are the rotation angle errors of rotary tables and the distance measurement error of the laser range finder. Using GUM algorithm, the uncertainties of these two sources are estimated and the system measurement uncertainty is evaluated. T...
Projects
Project (1)
A measurement instrument with non-orthogonal structure could save lots of time for manufacturers as it does not require to be assembled orthogonally and only needs one-time calibration. No part will be wasted because of orthogonal error within this framework. And these could be big advantages for manufacturers without highly advanced manufacture techniques. My current projects focus on developing different non-orthogonal instruments for industrial applications and the visual guiding strategies for these instruments to work automatically and precisely.
More relevant works written in Chinese can be found in following links:
1. Fengting Yang. Research on Non-orthogonal Measurement Instrument for Large-scale Metrology, 2017 (Master Thesis). http://kns.cnki.net/KCMS/detail/detail.aspx?dbcode=CMFD&dbname=CMFDTEMP&filename=1018061833.nh&uid=WEEvREcwSlJHSldRa1FhdXNXa0hIUUVEMFF0ODdVSWdGN000azREM1F4WT0=$9A4hF_YAuvQ5obgVAqNKPCYcEjKensW4IQMovwHtwkF4VYPoHbKxJw!!&v=MzIxNzRFYlBJUjhlWDFMdXhZUzdEaDFUM3FUcldNMUZyQ1VSTEtmWWVSbUZpM21XNzdNVkYyNkZyTytIOW5Qcko=
2. WU Bin,DING Wen,YANG Feng-ting,XUE Ting. The Error Analysis of the Non-orthogonal Total Station Coordinate Measurement System. Acta Metrologica Sinica, 2017, 38(6): 661-666. http://jlxb.china-csm.org/cn/abstract/abstract1105.shtml
3. Bin Wu, Shengwen Xie. Correction of the sight axis parameters for non-orthogonal laser theodolite, 光电子·激光 9(2015):1700-1706. http://www.joelcn.com/ch/reader/view_abstract.aspx?file_no=20150910
4. Bin, W. U., and S. Yang. "Research of measurement technology of non-orthogonal shaft laser theodolites." Laser Technology 39.5(2015):603-609. http://www.jgjs.net.cn/EN/Y2015/V39/I5/603
