Pith. sign in

REVIEW 1 cited by

Adaptive Bounding Box Uncertainties via Two-Step Conformal Prediction

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2403.07263 v2 pith:6PI5WN52 submitted 2024-03-12 cs.CV cs.LGstat.ML

Adaptive Bounding Box Uncertainties via Two-Step Conformal Prediction

classification cs.CV cs.LGstat.ML
keywords boundinguncertaintyconformalcoverageintervalsobjecttwo-stepadaptive
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
Share X Bluesky LinkedIn Reddit HN
read the original abstract

Quantifying a model's predictive uncertainty is essential for safety-critical applications such as autonomous driving. We consider quantifying such uncertainty for multi-object detection. In particular, we leverage conformal prediction to obtain uncertainty intervals with guaranteed coverage for object bounding boxes. One challenge in doing so is that bounding box predictions are conditioned on the object's class label. Thus, we develop a novel two-step conformal approach that propagates uncertainty in predicted class labels into the uncertainty intervals of bounding boxes. This broadens the validity of our conformal coverage guarantees to include incorrectly classified objects, thus offering more actionable safety assurances. Moreover, we investigate novel ensemble and quantile regression formulations to ensure the bounding box intervals are adaptive to object size, leading to a more balanced coverage. Validating our two-step approach on real-world datasets for 2D bounding box localization, we find that desired coverage levels are satisfied with practically tight predictive uncertainty intervals.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Instance-Level Post Hoc Uncertainty Quantification in Object Detection

    cs.CV 2026-06 unverdicted novelty 4.0

    Proposes MC-GLM to deliver instance-level post-hoc uncertainty for object detectors via Laplace approximation plus constant-cost Monte Carlo sampling.