Facebook’s AI Chief: Machines Could Learn Common Sense from Video
Google Cloud & YouTube-8M Video Understanding Challenge
Temporal Generative Adversarial Nets – The temporal generator consists of 1D deconvolutional layers and outputs a set of latent variables, each of which corresponds to a frame in the generated video, and the image generator transforms them into a video with 2D deconvolutional layers. This representation allows efficient training of the network parameters. Moreover, it can handle a wider range of applications including the generation of a long sequence, frame interpolation, and the use of pre-trained models. Experimental results demonstrate the effectiveness of our method.
Uses: Invariant feature learning, and Contrastive loss
Biologically-Plausible Unsupervised Learning of Face Representations from Natural Videos