Paper on RGB-D SLAM of Dr. ZHANG Yu's group accepted by TPAMI
"RGB-D SLAM in Dynamic Environments Using Point Correlations" (DSLAM), a paper of Dr. ZHANG Yu’s group has been accepted by IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI). The work focuses on the simultaneous localization and mapping (SLAM) in dynamic environment. It is the first paper accepted by TPAMI whose first author affiliation is the College of Control Science and Engineering, Zhejiang University. Moreover, it is also the first SLAM paper accepted by TPAMI whose first author affiliation is domestic university or research institution.In this paper, a simultaneous localization and mapping (SLAM) method that eliminates the influence of moving objects in dynamic environments is proposed. This method utilizes the correlation between map points to separate points that are part of the static scene and points that are part of different moving objects into different groups. A sparse graph is first created from all map points. In this graph, the vertices represent map points, and each edge represents the correlation between adjacent points. If the relative position between two points remains consistent over time, there is correlation between them, and they are considered to be moving together rigidly. If not, they are considered to have no correlation and to be in separate groups. After the edges between the uncorrelated points are removed, the remaining graph separates the map points of moving objects from the static scene. The largest group is assumed to be the group of reliable static map points. Experimental results demonstrate that robust and accurate performance can be achieved by the proposed SLAM method in both slightly and highly dynamic environments. Compared to other state-of-the-art methods, the proposed method can provide competitive accuracy with good real-time performance.To read the paper:https://ieeexplore.ieee.org/document/9145704Reporter: YU HongxiangEditor: WANG Jing
MORE >DMT clinches championship at 2020 iQIYI iCartoonFace Challenge
Under the umbrella of the College of Control Science and Engineering, a partnership between the the College and Alibaba DAMO Academy, a team of Zhejiang University students participated in and won the championship at the 2020 iQIYI iCartoonFace Challenge, an integral part of IJCAI PRICAI 2020.The team representing Zhejiang University was comprised of three students: HE Shuting (BA22), ZHANG Miao (BA21), and ZENG Zhaoyang (BA22). They came in first place in the iCartoonFace Recognition Challenge. This year iQIYI iCartoonFace Challenge attracted about 500 teams in China, such as Shanghai Jiao Tong University, the Chinese Academy of Sciences, the University of Science and Technology of China, Tencent and Dahua Technology. The DMT team is under the guidance of Dr. LUO Hao, Prof. JIANG Wei and Prof. LIU Xinggao, as well as researchers from Alibaba DAMO Academy. It has been actively engaging in several competitions in the field of computer vision and achieved excellent records. In the AI City Challenge 2020，DMT team innovatively integrated multi-domain learning and identity mining into vehicle city-Scale multi-Camera re-identification questions and was the third-place winner in the leader board of this track.This January the DMT team also participated in the 1st National Artificial Intelligence Challenge, co-hosted by Shenzhen Municipal Government and Shenzhen Tencent Information Technology Co., Ltd and won the third place in person re-identification track with a handsome sum of prize money (500,000 yuan).Reprinted from: ZJU Newsroom
MORE >
Add: 38 Zheda Rd.(Yuquan Campus) Hangzhou, Zhejiang 310027, P. R. China
Tel: +(00)86-571-87953386
Email: cse@zju.edu.cn