Buscar recursos académicos

In several applications, the trajectory of an entity, a feature or an object has to be tracked over a sequence of image frames. When the processing is to be performed in real time, there are important constraints leading...

We present Momentum Contrast (MoCo) for unsupervised visual representation learning. From a perspective on contrastive learning as dictionary look-up, we build a dynamic dictionary with a queue and a moving-averaged enco...