# Unsupervised part representation by Flow Capsules

Source: [Unsupervised part representation by Flow Capsules](<https://www.sota2.com/research/paper/68fa694d1e209131cea21597>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-28T13:56:46.474Z

Authors: Sara Sabour, Andrea Tagliasacchi, Soroosh Yazdani, Geoffrey E. Hinton, David J. Fleet

Published: 2020-11-27T18:59:42.000Z

## Abstract

Capsule networks aim to parse images into a hierarchy of objects, parts and relations. While promising, they remain limited by an inability to learn effective low level part descriptions. To address this issue we propose a way to learn primary capsule encoders that detect atomic parts from a single image. During training we exploit motion as a powerful perceptual cue for part definition, with an expressive decoder for part generation within a layered image model with occlusion. Experiments demonstrate robust part discovery in the presence of multiple objects, cluttered backgrounds, and occlusion. The part decoder infers the underlying shape masks, effectively filling in occluded regions of the detected shapes. We evaluate FlowCapsules on unsupervised part segmentation and unsupervised image classification.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2011.13920>)

[PDF](<https://arxiv.org/pdf/2011.13920.pdf>)