ELOS 2022

Feifei Gu

Feifei Gu, Speaker at Lasers Conferences
Shenzhen Institute of Advanced Technology, China
Title : Miniature and Accurate 3D Structured Light Sensor on Mobile Intelligent Terminal


3D vision technology has been recognized as one of the standard technologies for mobile intelligent terminals such as mobile phones to enter the new era of intelligent human-computer interaction. Now 3D vision technology has been successfully applied at mobile terminals in many fields, such as 3D live detection, virtual/augmented reality and so on. However, time-of-flight and laser speckle structured light technology are still the mainstream 3D vision technologies at mobile intelligent terminals nowadays, which were well known for their limited reconstruction accuracy. Therefore, both of them cannot be used to obtain accurate 3D information. In this case, their application scope at mobile terminals was seriously limited. To solve this problem, this work focused on the key issues of miniature structured light 3D accurate scanning method based on diffractive optical devices. By solving the key scientific problems such as high-density coding and high-precision 3D reconstruction under narrow baseline, high-precision and dynamic 3D imaging at mobile intelligent terminal can be realized. The main research contents are as follows: 1) one dimensional coding theory and method with small window and high capacity space based on epipolar constraint; 2) high precision detection and sub-pixel positioning method of the coding feature points; 3) high robust decoding algorithm of epipolar-coded pattern; 4) accurate 3D reconstruction method under narrow baseline condition. Based on the above researches, the imaging accuracy of 3D vision sensor at mobile terminals can be improved, and its application can be extended to 3D face accurate recognition, 3D vision guidance and other fields. In this sense, this research has important research value and application potential.

Audience take away:

  • People can know the current research progresses in 3D vision from my speech. Moreover, there may have potential 3D vision techniques that can be adopted in their researches.
  • In this work, a new 1D coding strategy based on epipolar homography was proposed, which may be applied in other field of pattern codification.
  • Miniature 3D structured light sensor was made by using diffractive optical devices, that may inspire other researchers to make their own 3D sensors.


Feifei Gu received a Ph.D. in Instrument Science and Technology from Xian Jiao Tong University in 2017. She is currently an associate research fellow at Shenzhen Institute of Advanced Technology (SIAT), Chinese Academy of Sciences (CAS). Her research interests include stereo vision-based 3D sensing and image processing. She has been the PI of over fifteen research grants including the Natural Science Foundation of China (NSFC), the Natural Science Foundation of Guangdong Province and so on. She has published more than 50 research papers, and has applied for over 16 patents until end of 2021.