A research team from Facebook AI conducts a large-scale study on unsupervised spatiotemporal representation learning from videos. The work takes a unified perspective on four recent image-based frameworks (MoCo, SimCLR, BYOL, SwAV) and investigates a simple objective that can easily generalize unsupervised representation learning methodologies to space-time.

Here is a quick read: Facebook AI Conducts Large-Scale Study on Unsupervised Spatiotemporal Representation Learning.

The paper A Large-Scale Study on Unsupervised Spatiotemporal Representation Learning is on arXiv.



Source link