CrossOver: 3D Scene Cross-Modal Alignment
This addresses the challenge of flexible cross-modal alignment in 3D scenes for applications like scene retrieval and object localization, though it appears incremental by building on existing multi-modal methods.
The paper tackles the problem of multi-modal 3D scene understanding by proposing CrossOver, a framework that aligns modalities like RGB images and point clouds without requiring complete data or rigid alignment, achieving superior performance on datasets such as ScanNet and 3RScan.
Multi-modal 3D object understanding has gained significant attention, yet current approaches often assume complete data availability and rigid alignment across all modalities. We present CrossOver, a novel framework for cross-modal 3D scene understanding via flexible, scene-level modality alignment. Unlike traditional methods that require aligned modality data for every object instance, CrossOver learns a unified, modality-agnostic embedding space for scenes by aligning modalities -- RGB images, point clouds, CAD models, floorplans, and text descriptions -- with relaxed constraints and without explicit object semantics. Leveraging dimensionality-specific encoders, a multi-stage training pipeline, and emergent cross-modal behaviors, CrossOver supports robust scene retrieval and object localization, even with missing modalities. Evaluations on ScanNet and 3RScan datasets show its superior performance across diverse metrics, highlighting the adaptability for real-world applications in 3D scene understanding.