Nan Xue. I'm a staff research scientist at Ant Research. My current research mainly lies on developing general learning algorithms for the computation of structured visual geometry in computer vision. Here is a research statement about my …

NEAT: Distilling 3D Wireframes from Neural Attraction Fields Nan Xue1 Bin Tan1,2 Yuxi Xiao1,3 Liang Dong4 Gui-Song Xia2 Tianfu Wu5* Yujun Shen1 1Ant Group 2Wuhan University 3Zhejiang University 4Google Inc. 5NC State University Image (DTU-23) Line3D++@LSD [12] Line3D++@HAWPv3 [12] LiMAP@HAWPv3 [17] NEAT (Ours)

N. Xue, S. Bai, F.-D. Wang, G.-S. Xia, T. Wu, L. Zhang, P.H.S. Torr IEEE Trans. on Pattern Analysis and Machine Intelligence ( TPAMI ), 2019. Learning Attraction Field Representation for Robust Line Segment Detection [ paper ]

Nan Xue ([email protected]) 1 Research Objectives Our 3D world is constructed by shapes, with humans innately skilled in observing and interpreting these shapes. We, as human beings, focus on the shape information, uncovering and utilizing the structural regularities within, and abstracting the world using geometric primitives. For instance, 3D

Nan Xue, Song Bai, Fu-Dong Wang, Gui-Song Xia, Tianfu Wu, Liangpei Zhang, Philip H.S. Torr Abstract—This paper presents regional attraction of line segment maps, and hereby poses the problem of line segment detection (LSD) as a problem of region coloring.

Nan Xue1,2, Song Bai3, Fudong Wang1, Gui-Song Xia1∗, Tianfu Wu 2, Liangpei Zhang1 1Wuhan University, China 2NC State University, USA 3University of Oxford, UK {xuenan, fudong-wang, guisong.xia, zlp62}@whu.edu.cn, [email protected], [email protected] Abstract This paper presents a region-partition based attraction

Nan Xue, Tianfu Wu, Song Bai, Fu-Dong Wang, Gui-Song Xia, Liangpei Zhang, Philip H.S. Torr Abstract—This article presents Holistically-Attracted Wireframe Parsing (HAWP), a method for geometric analysis of 2D images containing wireframes formed by line segments and junctions.

Zhucun Xue, Nan Xue, Gui-Song Xia∗, Weiming Shen CAPTAIN-LIESMARS, Wuhan University, Wuhan, 430079, China {zhucun.xue, xuenan, guisong.xia, shenwm}@whu.edu.cn Abstract This paper presents a new deep-learning based method to simultaneously calibrate the intrinsic parameters of fish-eye lens and rectify the distorted images. Assuming that

Fu-Dong Wang, Nan Xue, Student Member, IEEE, Yipeng Zhang, Gui-Song Xia, Senior Member, IEEE, and Marcello Pelillo, Fellow, IEEE Abstract—Graph matching is an important and persistent problem in computer vision and pattern recognition for finding node-to-node correspondence between graphs.

Fudong Wang 1, Nan Xue , Yipeng Zhang2, Xiang Bai3, and Gui-Song Xia1⋆ 1 State Key Lab. LIESMARS, Wuhan University, China {fudong-wang, xuenan, guisong.xia}@whu.edu.cn