Event is Evidence of the Growing Importance of AI
The first-ever virtual Computer Vision and Pattern Recognition (CVPR) Conference — one of the largest events exploring artificial intelligence (AI), machine learning, computer vision, deep learning, and more — successfully concluded last week. With more than 5,000 papers, keynotes, sessions, workshops, and tutorials amounting to 1,497,800 minutes of virtual learning and discussion, and an audience of 7,600 from more all over the world, the event served as a powerful forum to further the understanding, application, and advancement of computer vision, AI and machine learning on a global scale.
“We are very happy with the success of our first-ever CVPR Virtual despite the obvious challenges,” said Ramin Zabih, Cornell University Professor of Computer Science and Co-Chair of the CVPR 2020 Committee. “Thanks are due to the commitment, dedication, and determination of our volunteers, speakers, authors, and the larger computer vision community.”
Year after year, CVPR continues to grow as one of the largest and leading computer vision, AI, and machine learning events in the world. This year’s event saw record paper submissions, with 5,865, an increase of almost 30 percent over last year. In addition, CVPR 2020 featured high-profile industry leaders as keynote presenters during fireside chats. Satya Nadella, CEO of Microsoft delivered the main program’s opening keynote on 16 June, delivering a message of how computer vision can support and shape a post- COVID-19 environment. “This crisis has brought the need for digital technology and how it can help us, at large, as a society respond, recover and reimagine how we work and live going forward…” he said. “Digital technology, including computer vision, is going to play a big role in it… Once we come out of the COVID-19 crisis, we will use the flexibility to help people with not only their productivity but also their well-being, their needs.”
Charlie Bell, SVP, Amazon Web Services served as the closing keynote on 18 June. He shared an empowering message of the potential impact for computer vision, AI, machine learning on the future. “If you look at human history, it is very clear that as you get advances like this that improve human productivity, everyone benefits,” said Bell. “I believe we are at the dawn of a wonderful new era in human capability, and machine learning will make that possible. I think this will be a boon to human kind as we deploy this further.”
One of the event’s other prestigious components is its awards program. Each year, CVPR recognizes research through its best papers awards. This year, from a total of 1,467 accepted papers, which were whittled down to 26 finalists, the Awards Committee presented the following awards:
- Best Paper: Unsupervised Learning of Probably Symmetric Deformable 3D Objects from Images in the Wild
Authors: Shangzhe Wu (University of Oxford); Christian Rupprecht (University of Oxford); Andrea Vedaldi (University of Oxford)
- Best Student Paper: BSP-Net: Generating Compact Meshes via Binary Space Partitioning
Authors: Zhiqin Chen (Simon Fraser University); Andrea Tagliasacchi (Google Inc.); Hao Zhang (Simon Fraser University)
- Best Student Paper Honorable Mention: DeepCap: Monocular Human Performance Capture Using Weak Supervision
Authors: Marc Habermann (Max Plank Institute for Informatics); Weipeng Xu (MPII); Michael Zollhofer (Facebook Reality Labs); Gerard Pons-Moll (MPII, Germany); Christian Theobalt (MPI Informatik)
The Awards Committee also honored posthumously, Thomas S. Huang, who died on 25 April 2020. Huang was one of the leading figures in computer vision, pattern recognition, and human-computer interaction. In his honor, the Thomas Huang Memorial Prize has been established to recognize and honor distinguished individuals with long-standing service, research, and mentoring within the computer vision community. The prize will be awarded annually beginning in 2021.
“From the fireside chats and oral and poster sessions to the networking, engagement, and recognition opportunities, CVPR 2020 offered a unique forum for education, inspiration, and motivation,” said IEEE Computer Society Executive Director Melissa Russell, co-sponsor of the event. “To successfully execute an event of this magnitude and distinction within a fully virtual format is a testament to the ingenuity, strength, and passion of the computer vision community. We are honored to have been part of this event, and look forward to supporting next year’s conference.”
CVPR 2020 launched 14 June and concluded 19 June. A post-event report with additional program insights, paper and author highlights, sponsor details, and future trends will be made available in the coming weeks. CVPR 2021 will take place 21 June – 24 June in Nashville, Tennessee.