Profile
I am a Master's student at University of Alberta. I work in the Vision and Robotics lab under supervision of Dr. Martin Jagersand. My research interest is focused on 3D reconstruction and SLAM (Simultaneous Localization and Mapping) in Computer Vision. My thesis project is titled "Incremental 3D Line Segment Extraction for Surface Reconstruction from Semi-dense SLAM". I am also interested in 2D Computer Vision problems including object tracking, segmentation and recognition.
Projects
- It is challenging to utilize the large scale point clouds of semi-dense SLAM for real-time surface reconstruction. In order to obtain meaningful surfaces and reduce the number of points used in surface reconstruction, we propose to simplify the point clouds generated by semi-dense SLAM using 3D line segments. Specifically, we present a novel incremental approach for real-time 3D line segments extraction. Our experimental results show that the 3D line segments generated by our method are highly accurate compared to other methods. With the reconstructed surfaces, we demonstrate that using the extracted 3D line segments greatly improves the quality of 3D surface compared to using the 3D points directly from SLAM systems.
- We developed a 3D reconstruction pipeline that can automatically remove and replace 3D objects in the reconstructed scene. Using deep neural networks, we perform object segmentation on 2D images. With the objects identified, we are able to segment out the 3D objects in the 3D scene. After analyzing the objects' position and orientation, replacing the objects can be achieved. Through a simple web interface, user can upload a video and, for example, reconstruct a room and see how it looks with a different sofa.
Experiences
When I was working for Pair as an intern, I surveyed the field of 3D reconstruction and developed a 3D reconstruction pipeline that can automatically remove and replace 3D objects in the reconstructed scene.
I was the teaching assistant for course CMPUT201 (Practical Programming Methodology) and CMPUT301 (Introduction to Software Engineering). For CMPUT201, I graded assignments, gave short lectures in the lab and answered questions in the lab and discussion forum. For CMPUT301, besides all the regular TA duties, I was also the project manager of student teams, helped them design and develop their Android applications.
I worked at the registration desk. Also, we presented a demo in the KUKA booth.
Publications
[1] Incremental 3D Line Segment Extraction from Semi-dense SLAM, Shida He, Xuebin Qin, Zichen Zhang and Martin Jagersand, accepted at International Conference on Pattern Recognition (ICPR), August 2018.[2] ByLabel: A Boundary based Semi-Automatic Image Annotation Tool, Xuebin Qin, Shida He, Zichen Zhang, Masood Dehghan and Martin Jagersand, IEEE Winter Conf. on Applications of Computer Vision (WACV), March 2018.
[3] Real-time salient closed boundary tracking using perceptual grouping and shape priors, Xuebin Qin, Shida He, Zichen Zhang, Masood Dehghan and Martin Jagersand, British Machine Vision Conference (BMVC), September 2017.
[4] Real-time salient closed boundary tracking via line segments perceptual grouping, Xuebin Qin, Shida He, Camilo Perez Quintero, Abhineet Singh, Masood Dehghan and Martin Jagersand, IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), September 2017.