Differentiable Mathematical Programming for Object-Centric Representation Learning

TL;DR

This paper introduces a differentiable, topology-aware graph cut method for object-centric representation learning, improving scalability and performance in object discovery tasks with textured scenes.

Contribution

It presents a novel differentiable graph cut approach for object-centric learning that explicitly encodes neighborhood relationships and is more scalable than previous methods.

Findings

Outperforms existing methods on textured scene object discovery

Scalable and efficient quadratic programming approximation

Explicitly encodes neighborhood relationships in image graphs

Abstract

We propose topology-aware feature partitioning into $k$ disjoint partitions for given scene features as a method for object-centric representation learning. To this end, we propose to use minimum $s$-$t$ graph cuts as a partitioning method which is represented as a linear program. The method is topologically aware since it explicitly encodes neighborhood relationships in the image graph. To solve the graph cuts our solution relies on an efficient, scalable, and differentiable quadratic programming approximation. Optimizations specific to cut problems allow us to solve the quadratic programs and compute their gradients significantly more efficiently compared with the general quadratic programming approach. Our results show that our approach is scalable and outperforms existing methods on object discovery tasks with textured scenes and objects.