Organizer: Cyberspace Security Institute
Report time:202May 28, 2004, 2:0 p.m0-4:00
Presentation venue: Tencent Conference号:166-942-633
Speaker: Wang Longguang
Report content:
With the rapid development of sensors such as lidar and depth cameras, 3D data acquisition is becoming more and more convenient,其It has been widely used in the fields of automatic driving, robotics, etc., for the perception and understanding of three-dimensional scenes and modeling transmission。However, when the scene scale continues to expand, the understanding, modeling and compression of three-dimensional scenes面临Challenges such as high tagging cost, slow processing speed, coarse modeling granularity, and large storage overhead。In response to the above challenges, this report focuses on understanding, modeling, and compression of 3D scenesSolution strategy and path。
Brief introduction of reporter:
Dr. Wang Longguang,His research interests include low-level computer vision and 3D visionPapers published in IEEE TPAMI, IEEE TIP, CVPR, ICCV, ECCV 30余One of them11 CCF Class A papers,ESIHot papers2 articles, ESIHighly cited papers1 article, Google academic total citations more than 3100 times。Presided over the National Natural Science Foundation project,入选2023global2%List of scientists。Was in the position ofPRCV 2023 Chair in the Field of Underlying Vision and guest editor of journals such as IET Computer Vision;Have organizedCVPR 2022/2023 Binocular Image/Light Field Image Super Resolution Challenge, VALSE 2023 "3D Vision Frontier Technology" seminar。担任Reviewer for journals such as IEEE TPAMI, IJCV, IEEE TIP, and conferences such as CVPR, ICCV, ECCV。(文/Liu Xige 审/ Yang Cheng, Song Wei, Huang Lijuan)