Meta AI Extends MAEs to Video for Self-Supervised Representation Learning With Minimal Domain Knowledge

In the new paper Masked Autoencoders As Spatiotemporal Learners, a Meta AI research team extends masked autoencoders (MAE) to spatiotemporal representation learning for video. The novel approach introduces negligible inductive biases on space-time while achieving strong empirical results compared to vision transformers (ViTs) and outperforms supervised pretraining by large margins.

ai artificial intelligence deep-neural-networks domain knowledge knowledge learning machine learning machine learning & data science masked autoencoder meta meta ai ml representation representation learning research self-supervised learning technology video