Masked Multi-Query Slot Attention for Unsupervised Object Discovery

We introduce a sequential extension to Slot Attention — Slot Attention for Video — which we train to predict frame reconstructions or optical flow

Slot Attention is an architectural component that interfaces with perceptual representations such as the output of a convolutional neural network and

slot com This module is built on top of the recently-emerged slot attention , which offers an object-centric approach for image representation The xSlot module

slot hopper videos Implementation of Slot Attention from the paper 'Object-Centric Learning with Slot Attention' in Pytorch Here is a video that describes what this network