Reg3D: Reconstructive Geometry Instruction Tuning for 3D Scene Understanding
By: Hongpei Zheng , Lintao Xiang , Qijun Yang and more
Potential Business Impact:
Helps computers understand 3D spaces like we do.
The rapid development of Large Multimodal Models (LMMs) has led to remarkable progress in 2D visual understanding; however, extending these capabilities to 3D scene understanding remains a significant challenge. Existing approaches predominantly rely on text-only supervision, which fails to provide the geometric constraints required for learning robust 3D spatial representations. In this paper, we introduce Reg3D, a novel Reconstructive Geometry Instruction Tuning framework that addresses this limitation by incorporating geometry-aware supervision directly into the training process. Our key insight is that effective 3D understanding necessitates reconstructing underlying geometric structures rather than merely describing them. Unlike existing methods that inject 3D information solely at the input level, Reg3D adopts a dual-supervision paradigm that leverages 3D geometric information both as input and as explicit learning targets. Specifically, we design complementary object-level and frame-level reconstruction tasks within a dual-encoder architecture, enforcing geometric consistency to encourage the development of spatial reasoning capabilities. Extensive experiments on ScanQA, Scan2Cap, ScanRefer, and SQA3D demonstrate that Reg3D delivers substantial performance improvements, establishing a new training paradigm for spatially aware multimodal models.
Similar Papers
Ross3D: Reconstructive Visual Instruction Tuning with 3D-Awareness
CV and Pattern Recognition
Teaches computers to understand 3D spaces from many pictures.
Seeing 3D Through 2D Lenses: 3D Few-Shot Class-Incremental Learning via Cross-Modal Geometric Rectification
CV and Pattern Recognition
Teaches computers to recognize 3D objects better.
Think with 3D: Geometric Imagination Grounded Spatial Reasoning from Limited Views
CV and Pattern Recognition
Lets computers imagine 3D shapes from pictures.