Simplified priors for Object-Centric Learning

TL;DR

This paper introduces SAMP, a simple, fully-differentiable, scalable object-centric learning method that uses convolutional and max pooling layers to create primitive slots, outperforming previous complex methods on benchmarks.

Contribution

The paper presents SAMP, a novel, simple, and scalable object-centric learning approach that relies solely on convolutional and max pooling layers, eliminating the need for complex or non-differentiable components.

Findings

SAMP outperforms previous methods on standard benchmarks.

SAMP is fully differentiable and scalable.

SAMP uses only convolutional, max pooling, and attention layers.

Abstract

Humans excel at abstracting data and constructing \emph{reusable} concepts, a capability lacking in current continual learning systems. The field of object-centric learning addresses this by developing abstract representations, or slots, from data without human supervision. Different methods have been proposed to tackle this task for images, whereas most are overly complex, non-differentiable, or poorly scalable. In this paper, we introduce a conceptually simple, fully-differentiable, non-iterative, and scalable method called SAMP Simplified Slot Attention with Max Pool Priors). It is implementable using only Convolution and MaxPool layers and an Attention layer. Our method encodes the input image with a Convolutional Neural Network and then uses a branch of alternating Convolution and MaxPool layers to create specialized sub-networks and extract primitive slots. These primitive slots are then used as queries for a Simplified Slot Attention over the encoded image. Despite its simplicity, our method is competitive or outperforms previous methods on standard benchmarks.