Adaptive Geodesic Conformal Prediction for Egocentric Camera Pose Estimation

TL;DR

This paper introduces an adaptive conformal prediction method for egocentric camera pose estimation that guarantees uncertainty regions and effectively identifies harder frames to improve coverage in AR applications.

Contribution

It proposes DINOv2-Bridge, a two-stage adaptive conformal prediction approach that transfers difficulty estimation across participants without additional images, enhancing coverage guarantees.

Findings

Standard CP achieves 90% overall coverage but only 60% on hardest frames.

Geodesic scoring better identifies physically harder frames than Euclidean scoring.

DINOv2-Bridge improves hardest frame coverage from 75% to 93% while maintaining overall 90% coverage.

Abstract

Egocentric pose estimation for Augmented Reality (AR) and assistive devices requires not just accurate predictions but guaranteed uncertainty regions. Conformal prediction (CP) provides such guarantees without retraining, but we show that standard CP with a single fixed threshold achieves nominal 90% overall coverage while covering only ~60% of the hardest 25% of frames (Q4) -- a ~30 percentage-point conditional coverage gap consistent across 12 participants, 3 predictors, and 3 horizons (108 evaluations) on EPIC-Fields. We further show that a geodesic SE(3) nonconformity score identifies physically harder frames than Euclidean scoring, with only 15-26% Q4 overlap and 2-3x higher ground-truth camera displacement for geodesic Q4 frames. To close the coverage gap, we propose DINOv2-Bridge adaptive CP: a two-stage difficulty estimator trained on a single source participant that transfers cross-participant without any images at test time, improving Q4 coverage from ~0.75 to ~0.93 while maintaining overall coverage at the 90% target.