Vision-Language Pre-training with Object Contrastive Learning for 3D Scene Understanding
This addresses the problem of generalizing vision-language models to 3D scene understanding for applications in robotics and augmented reality, but it is incremental as it adapts existing 2D methods to 3D data.
The paper tackles the lack of universal 3D vision-language embeddings for point cloud data by proposing 3DVLP, a pre-training framework with object contrastive learning, which achieves excellent performance on three 3D vision-language tasks.
In recent years, vision language pre-training frameworks have made significant progress in natural language processing and computer vision, achieving remarkable performance improvement on various downstream tasks. However, when extended to point cloud data, existing works mainly focus on building task-specific models, and fail to extract universal 3D vision-language embedding that generalize well. We carefully investigate three common tasks in semantic 3D scene understanding, and derive key insights into the development of a pre-training model. Motivated by these observations, we propose a vision-language pre-training framework 3DVLP (3D vision-language pre-training with object contrastive learning), which transfers flexibly on 3D vision-language downstream tasks. 3DVLP takes visual grounding as the proxy task and introduces Object-level IoU-guided Detection (OID) loss to obtain high-quality proposals in the scene. Moreover, we design Object-level Cross-Contrastive alignment (OCC) task and Object-level Self-Contrastive learning (OSC) task to align the objects with descriptions and distinguish different objects in the scene, respectively. Extensive experiments verify the excellent performance of 3DVLP on three 3D vision-language tasks, reflecting its superiority in semantic 3D scene understanding.