With the development of deep learning theory and technology, the performance of computer vision algorithms including object detection and tracking, face recognition, and 3D reconstruction have made tremendous progress. However, computer vision technology relies on the valid information from the input image and video, and the performance of the algorithm is essentially constrained by the quality of source image/video.
Gigapixel videography plays important role in capturing large-scale dynamic scenes for both macro and micro domains. Benefited from the recent progress of gigapixel camera design, the capture of gigapixel-level image/video becomes more and more convenient. In particular, along with the emergence of gigapixel-level image/video, the corresponding computer vision tasks remain unsolved, due to the extremely high-resolution, large-scale, huge-data that induced by the gigapixel camera.
GigaVision is the workshop of computer vision in gigapixel videography, aiming to introduce a latest video/image dataset with gigapixel-level resolution and high dynamic range in terms of spatial, temporal, angle and spectrum dimensions and promote the application of such big data in computer vision tasks.
The workshop invites submissions of original high-quality contributions. Relevant work that has been recently published, is in progress, or is to be presented other venues including the CVPR main conference is also welcome. Topics of interests include, but are not limited to:
|13:00-13:10||Welcome and Opening Remarks|
|13:10-13:50||Keynote Talk 1 Title: Learning to Track and Segment Multiple Objects in Videos Speaker: Ming-Hsuan Yang, University of California at Merced, USA|
|13:50-14:30||Keynote Talk 2 Title: Precision Modeling of 3D Human Motion for Behavioural and Performance Analysis Speaker: Ajmal Mian, The University of Western Australia, Australia|
|14:30-15:00||Invited Talk 1 Title: Visual Recognition with Knowledge (VR-K): from an Active Agent's Perspective Speaker: Yezhou Yang, Arizona State University, USA|
|15:00-15:30||Invited Talk 2 Title: Gaussian Attacks on the Deep Neural Networks with High-Resolution inputs in A Low-Dimensional Space by Learning the Distribution of Adversarial Examples Speaker: Boqing Gong, Google, USA|
|16:00-16:30||Invited Talk 3 Title: Multi-person articulated tracking framework Speaker: Chen Qian, SenseTime, China|
|16:30-17:00||Invited Talk 4 Title: Towards Weakly-Supervised Visual Scene Understanding Speaker: Zhiding Yu, NVIDIA, USA|
|17:00-17:45||Invited Paper Presentation Paper: AdaFrame: Adaptive Frame Selection for Fast Video Recognition Speaker: Zuxuan Wu, University of Maryland, USA Paper: Adaptive NMS: Refining Pedestrian Detection in a Crowd Speaker: Songtao Liu, Beihang University, China Paper: Collaborative Global-Local Networks for Memory-Efficient Segmentation of Ultra-High Resolution Images Speaker: Wuyang Chen, Texas A&M University, USA|