Carl Vondrick

TL;DR: A generative adversarial network for video with a spatio-temporal convolutional architecture that untangles the scene's foreground from the background is proposed that can generate tiny videos up to a second at full frame rate better than simple baselines.

TL;DR: The AVA dataset densely annotates 80 atomic visual actions in 437 15-minute video clips, where actions are localized in space and time, resulting in 1.59M action labels with multiple labels per person occurring frequently.

TL;DR: In this article, the authors leverage the natural synchronization between vision and sound to learn an acoustic representation using two-million unlabeled videos and propose a student-teacher training procedure which transfers discriminative visual knowledge from well established visual recognition models into the sound modality using unlabelled video as a bridge.

TL;DR: The authors proposed a generative adversarial network for video with a spatio-temporal convolutional architecture that untangles the scene's foreground from the background, which can generate tiny videos up to a second at full frame rate better than simple baselines.

TL;DR: A new large-scale video dataset designed to assess the performance of diverseVisual event recognition algorithms with a focus on continuous visual event recognition (CVER) in outdoor areas with wide coverage is introduced.