Map-Det3D

Map-Det3D: Metric Feed-Forward 3D Reconstruction Prior for Multi-view 3D Object Detection from Streaming Inputs

ECCV 2026

Yung-Hsu Yang1      Luigi Piccinelli1      Samuel Rota Bulò2      Sunghwan Hong1      Denis Rozumny2     
Johannes Schönberger2      Zuria Bauer1      Hermann Blum3      Peter Kontschieder2      Marc Pollefeys1

1ETH Zürich     2Meta Reality Labs Zürich     3University of Bonn

Abstract

Metric 3D object detection is a core capability for embodied agents, yet most reliable systems lean on depth sensors, trading away cost, power, and integration simplicity. This motivates monocular 3D detection, which avoids additional constraints, yet it faces a major obstacle: from a single image, depth, and especially absolute scale, are underconstrained. As a result, the prevailing pattern of detecting in 2D and then predicting 3D attributes is often brittle, since modest range errors can dominate 3D localization, and the learned scale prior can fail when cameras, motion, or environments undergo domain shifts. To address this, we propose Map-Det3D, an online multi-view 3D object detection model that brings detection directly into a 3D space reconstructed from RGB. We map a short temporal window into multiple views and repurpose a feed-forward metric 3D reconstruction model as our geometric backbone while tuning its object-aware capabilities. Building on this representation, Map-Det3D directly predicts boxes in metric 3D space, without the widely used 2D-to-3D lifting. Experiments across different benchmarks show that this design supports strong online performance and robust transfer without adaptation, suggesting that training reconstruction priors for detection is a practical route to stable metric 3D detection from monocular video.


Summary

  • Map-Det3D is an RGB-only, online multi-view indoor 3D detector that brings metric reconstruction priors into a detection transformer.
  • Its direct 3D box head uses FF3R to recover metric scale, avoiding error-prone 2D-to-3D lifting.
  • Component studies validate direct 3D prediction, temporal aggregation, and camera conditioning, alongside strong cross-benchmark generalization.

BibTeX


Awesome webpage template