Confident Object Detection via Conformal Prediction and Conformal Risk Control: an Application to Railway Signaling

Léo Andéol, Thomas Fel, Florence De Grancey, Luca Mossina
4/12/2023
cs.LGstat.ML

No official code found. Best implementation: leoandeol/conformal_railway_signal_detection (13 stars, MIT, last commit 2023-05-08).

Abstract

Deploying deep learning models in real-world certified systems requires the ability to provide confidence estimates that accurately reflect their uncertainty. In this paper, we demonstrate the use of the conformal prediction framework to construct reliable and trustworthy predictors for detecting railway signals. Our approach is based on a novel dataset that includes images taken from the perspective of a train operator and state-of-the-art object detectors. We test several conformal approaches and introduce a new method based on conformal risk control. Our findings demonstrate the potential of the conformal prediction framework to evaluate model performance and provide practical guidance for achieving formally guaranteed uncertainty bounds.

View on arXivView PDF

Code Implementations(1)

leoandeol/conformal_railway_signal_detectionCommunity63%
MIT

Confident Object Detection via Conformal Prediction and Conformal Risk Control: an Application to Railway Signaling

131Python, Jupyter NotebookMay 1, 20233 years agoMIT

Similar Papers

Conformal Object Detection by Sequential Risk Control

86% match2025

Onboard Out-of-Calibration Detection of Deep Learning Models using Conformal Prediction

84% match2024

Conformal Prediction and Trustworthy AI

83% match2025

Adaptive Bounding Box Uncertainties via Two-Step Conformal Prediction

82% match1 repo132026

Empirically Validating Conformal Prediction on Modern Vision Architectures Under Distribution Shift and Long-tailed Data

82% match2023

Quantifying Deep Learning Model Uncertainty in Conformal Prediction

81% match2023

Cite this paper

@article{andol2023confident,
  title  = {Confident Object Detection via Conformal Prediction and Conformal Risk Control: an Application to Railway Signaling},
  author = {Léo Andéol and Thomas Fel and Florence De Grancey and Luca Mossina},
  year   = {2023},
  eprint = {2304.06052},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2304.06052v2}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2304.06052.svg)](https://codeofpaper.com/paper/2304.06052)

Discussion