PARTICUL: Part Identification with Confidence measure using Unsupervised Learning

TL;DR

PARTICUL introduces an unsupervised method for detecting object parts in fine-grained recognition datasets, leveraging macro-similarities and confidence measures to improve interpretability and classification performance.

Contribution

It presents a novel unsupervised algorithm with confidence estimation for part detection, enhancing interpretability in fine-grained recognition tasks.

Findings

Effective part detection on Caltech-UCSD Bird 200 and Stanford Cars datasets.

Detectors provide reliable confidence measures for part visibility.

Improved interpretability with competitive classification accuracy.

Abstract

In this paper, we present PARTICUL, a novel algorithm for unsupervised learning of part detectors from datasets used in fine-grained recognition. It exploits the macro-similarities of all images in the training set in order to mine for recurring patterns in the feature space of a pre-trained convolutional neural network. We propose new objective functions enforcing the locality and unicity of the detected parts. Additionally, we embed our detectors with a confidence measure based on correlation scores, allowing the system to estimate the visibility of each part. We apply our method on two public fine-grained datasets (Caltech-UCSD Bird 200 and Stanford Cars) and show that our detectors can consistently highlight parts of the object while providing a good measure of the confidence in their prediction. We also demonstrate that these detectors can be directly used to build part-based fine-grained classifiers that provide a good compromise between the transparency of prototype-based approaches and the performance of non-interpretable methods.