PARTICUL: Part Identification with Confidence measure using Unsupervised Learning
It addresses the problem of interpretable part detection in fine-grained recognition for computer vision researchers, offering a compromise between transparency and performance, though it is incremental as it builds on existing unsupervised and pre-trained methods.
The paper tackles unsupervised learning of part detectors for fine-grained recognition by mining recurring patterns in pre-trained CNN features, achieving consistent part highlighting and a confidence measure for part visibility on Caltech-UCSD Bird 200 and Stanford Cars datasets.
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.