CM-EVS: Sparse Panoramic RGB-D-Pose Data for Complete Scene Coverage

2026-05-15

AI Summary

Existing 3D assets often lack sparse, geometrically consistent, and comparable panoramic training interfaces, resulting in viewpoint redundancy, heterogeneous annotations, or insufficient scene coverage. To address this, this work proposes COVER, a training-agnostic viewpoint selection strategy that leverages equirectangular projection (ERP) probes, depth-range mapping, and a geometric conflict penalty mechanism. By integrating incremental coverage scoring with greedy optimization, COVER achieves near-optimal scene coverage while bounding proxy error. Using this approach, the authors construct the CM-EVS dataset, comprising 36,373 indoor and outdoor panoramic RGB-D-pose frames. Remarkably, an average of only 25 frames per scene suffices to cover 13 room categories, significantly reducing redundancy while striking an improved balance between coverage completeness and geometric consistency.

Abstract

Modern 3D visual learning relies on observations sampled from metric 3D assets, yet existing scans, meshes, point clouds, simulations, and reconstructions do not directly provide a sparse, comparable, and geometry-consistent panoramic training interface. Dense trajectories duplicate nearby views, source-specific rendering policies yield heterogeneous annotations, and sparse heuristics may miss important regions or introduce depth-inconsistent observations. We study how to convert 3D assets into sparse panoramic RGB-D-pose data that preserves complete scene coverage with low redundancy and auditable provenance. We propose COVER (Coverage-Oriented Viewpoint curation with ERP Range-depth warping), a training-free ERP viewpoint curator that projects geometry observed from selected views into candidate ERP probes, scores incremental coverage, and penalizes depth conflicts. Under bounded proxy error, its greedy coverage proxy preserves the standard coverage-style approximation behavior up to an additive error term. Using COVER, we build CM-EVS (Coverage-curated Metric ERP View Set), a panoramic RGB-D-pose dataset with 36,373 curated ERP frames from 1,275 indoor scenes across Blender indoor, HM3D, and ScanNet++, complemented by outdoor panoramas from TartanGround and OB3D re-encoded into the same schema. Each frame provides full-sphere RGB, metric range depth, calibrated pose; COVER-produced indoor frames include per-step provenance logs. With a median of only 25 frames per indoor scene, CM-EVS covers all 13 unified room types while maintaining compact scene-level coverage. Experiments show that COVER improves the coverage-conflict trade-off, making CM-EVS a sparse, compact, and auditable RGB-D-pose resource for geometry-consistent panoramic 3D learning.

Problem

• sparse panoramic data
• complete scene coverage
• RGB-D-pose
• geometry consistency
• 3D visual learning

Innovation

• sparse panoramic data
• geometry-consistent coverage
• ERP viewpoint curation
• RGB-D-pose dataset
• depth conflict penalization