cs.CVNov 2, 2026

ReVA-HPE: Relative 6DoF Head Pose Estimation with Synthetic Pose Supervision and Personalized Virtual Anchors

Authors: Vasiliki Vasileiou, Panagiotis P. Filntisis, Petros Maragos, Kostas Daniilidis

Organizations: HERON - Hellenic Robotics Center of Excellence, Athena Research Center, Greece · National Technical University of Athens, School of Electrical and Computer Engineering, Greece · Robotics Institute, Athena Research Center, Greece · University of Pennsylvania, USA

Abstract

Monocular head pose estimation is usually formulated as absolute regression from a single face crop, yet many applications observe the same person over time and can supply an explicit anchor image. Recently, relative head pose estimation was proposed, which casts the problem as estimation of the relative geometric change between an anchor and a query image. However, existing methods still require the anchor image pose to be known at inference, significantly limiting their practical use. We first extend relative head pose estimation to full 6DoF by additionally predicting a depth-normalized relative translation, which composes with the anchor to recover metric query pose in the camera frame. To eliminate the need for a ground-truth anchor pose, we reconstruct a personalized avatar from a single unlabeled enrollment image and render a bank of virtual anchors with known poses. The bank provides robust rotation hypotheses through multi-anchor consensus, while the real enrollment image serves as the real anchor in the translation branch via landmark PnP. We also introduce an efficient variant that amortizes the bank cost once at enrollment, after which each query requires a single real-to-real relative prediction, with limited accuracy sacrifice. The relative pose estimator is trained exclusively on synthetic face renderings, without real pose supervision, pseudo-labels, domain adaptation, or fine-tuning on the evaluation sets. On real 6DoF benchmarks, ReVA-HPE achieves state-of-the-art rotation accuracy on BIWI and, given a near-frontal enrollment image, on FaMoS, while remaining competitive in metric translation despite relying solely on synthetic pose supervision. Project page and code: https://vasilikivas.github.io/ReVA-HPE/

Figures & tables

Explore similar work

CardsList
  1. VGGT-HPE: Reframing Head Pose Estimation as Relative Pose Prediction

    Apr 11, 2026Vasiliki Vasileiou, Panagiotis P. Filntisis, Petros Maragos +1Relative Pose EstimationObject Pose Estimation

  2. Egocentric Visibility-Aware Human Pose Estimation

    Feb 27, 2026Peng Dai, Yu Zhang, Yiqiang Feng +2Human Pose Estimation

  3. Prior-free relative 6D pose estimation of multiple object instances

    Sep 8, 2026Behdad Khodabandehloo, Andrea Caraffa, Davide Boscaini +1Relative Pose EstimationObject Pose Estimation