What could go wrong? Discovering and describing failure modes in computer vision

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Csurka, Gabriela, Hayes, Tyler L., Larlus, Diane, Volpi, Riccardo
Format: Preprint
Publié: 2024
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866912041887858688
author Csurka, Gabriela
Hayes, Tyler L.
Larlus, Diane
Volpi, Riccardo
author_facet Csurka, Gabriela
Hayes, Tyler L.
Larlus, Diane
Volpi, Riccardo
contents Deep learning models are effective, yet brittle. Even carefully trained, their behavior tends to be hard to predict when confronted with out-of-distribution samples. In this work, our goal is to propose a simple yet effective solution to predict and describe via natural language potential failure modes of computer vision models. Given a pretrained model and a set of samples, our aim is to find sentences that accurately describe the visual conditions in which the model underperforms. In order to study this important topic and foster future research on it, we formalize the problem of Language-Based Error Explainability (LBEE) and propose a set of metrics to evaluate and compare different methods for this task. We propose solutions that operate in a joint vision-and-language embedding space, and can characterize through language descriptions model failures caused, e.g., by objects unseen during training or adverse visual conditions. We experiment with different tasks, such as classification under the presence of dataset bias and semantic segmentation in unseen environments, and show that the proposed methodology isolates nontrivial sentences associated with specific error causes. We hope our work will help practitioners better understand the behavior of models, increasing their overall safety and interpretability.
format Preprint
id arxiv_https___arxiv_org_abs_2408_04471
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle What could go wrong? Discovering and describing failure modes in computer vision
Csurka, Gabriela
Hayes, Tyler L.
Larlus, Diane
Volpi, Riccardo
Computer Vision and Pattern Recognition
Deep learning models are effective, yet brittle. Even carefully trained, their behavior tends to be hard to predict when confronted with out-of-distribution samples. In this work, our goal is to propose a simple yet effective solution to predict and describe via natural language potential failure modes of computer vision models. Given a pretrained model and a set of samples, our aim is to find sentences that accurately describe the visual conditions in which the model underperforms. In order to study this important topic and foster future research on it, we formalize the problem of Language-Based Error Explainability (LBEE) and propose a set of metrics to evaluate and compare different methods for this task. We propose solutions that operate in a joint vision-and-language embedding space, and can characterize through language descriptions model failures caused, e.g., by objects unseen during training or adverse visual conditions. We experiment with different tasks, such as classification under the presence of dataset bias and semantic segmentation in unseen environments, and show that the proposed methodology isolates nontrivial sentences associated with specific error causes. We hope our work will help practitioners better understand the behavior of models, increasing their overall safety and interpretability.
title What could go wrong? Discovering and describing failure modes in computer vision
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2408.04471