UT Dallas > Computer Science > Award > Dr. Xiaohu Guo and his Collaborators Awarded the ACM SIGGRAPH 2023 Technical Papers Best Paper Award

Dr. Xiaohu Guo and his Collaborators Awarded the ACM SIGGRAPH 2023 Technical Papers Best Paper Award

This August, Dr. Xiaohu Guo, his PhD student Ms. Ningna Wang, and a team of collaborators received the Association for Computing Machinery (ACM) Special Interest Group on Computer GRAPHics and Interactive Techniques (SIGGRAPH)’s Technical Papers Best Paper Award for the paper titled “Globally Consistent Normal Orientation for Point Clouds by Regularizing the Winding Number Field.” ACM SIGGRAPH is a special interest group within ACM, and SIGGRAPH 2023 is the premier conference for computer graphics and interactive techniques worldwide. The ACM SIGGRAPH community is a global nonprofit organization serving the evolution of computer graphics and interactive techniques. With thousands of members across the world, the researchers, artists, developers, filmmakers, scientists, and business professionals of ACM SIGGRAPH are building the future of digital art and interactive design.

“3D scanning is an essential technology supporting the foundational framework of Augmented Reality (AR), Digital Twins, and Metaverse. Across all 3D scanning methods, their collected raw data usually manifest as 3D point clouds. To obtain seamless 3D models – such as the 3D polygonal meshes commonly employed in video games or VR/AR applications – a procedure named as surface reconstruction becomes essential for transforming those initial point clouds into watertight surfaces,” Dr. Guo said. “Our paper introduces a critical technology for surface reconstruction.”

“A successful surface reconstruction algorithm relies on the requirement that all the normal directions of points are oriented outwards consistently and globally, which is a very tricky condition to satisfy when the point clouds are scanned from real-world objects, such as those with intricate topology, sharp features, thin structures, or data imperfections due to limitations in the scanning technology. This paper introduces an innovative algorithm designed to ensure the consistent orientation of normal directions within point clouds. This is achieved by regulating the winding number field of the reconstructed surface, ultimately leading to the successful and precise reconstruction of surfaces while maintaining resilience against various challenges,” he concluded.

Dr. Guo and Ms. Wang collaborated on this project with a team of computer scientists: Mr. Rui Xu, Dr. Shiqing Xin, Dr. Mingyan Jiang, and Dr. Changhe Tu at Shandong University, Mr. Zhiyang Dou at the University of Hong Kong, Dr. Shuangmin Chen at Qingdao University of Science and Technology, and Dr. Wenping Wang at Texas A&M University. 

This year marked the 50th anniversary of the SIGGRAPH conference. This year’s conference drew one of the most substantial audiences, with approximately 20,000 people in attendance. There were 616 technical paper submissions from 25 countries, of which 126 were accepted as Journal Track Papers and 86 Conference Track Papers. The Journal Track Papers are published in ACM Transactions on Graphics. Five of the 126 journal track papers were selected for the Technical Papers Best Paper Award. Dr. Guo’s paper was one of those five papers chosen.