Open Vocabulary 3D Scene Understanding via Geometry Guided Self-Distillation
This addresses the problem of limited 3D-text data for researchers and practitioners in 3D vision, offering a novel method that improves over prior distillation-based approaches.
The paper tackles the challenge of open vocabulary 3D scene understanding by proposing Geometry Guided Self-Distillation (GGSD), which leverages 2D pre-trained models and 3D geometric priors to enhance 3D representations, resulting in performance that significantly surpasses existing methods on benchmark datasets.
The scarcity of large-scale 3D-text paired data poses a great challenge on open vocabulary 3D scene understanding, and hence it is popular to leverage internet-scale 2D data and transfer their open vocabulary capabilities to 3D models through knowledge distillation. However, the existing distillation-based 3D scene understanding approaches rely on the representation capacity of 2D models, disregarding the exploration of geometric priors and inherent representational advantages offered by 3D data. In this paper, we propose an effective approach, namely Geometry Guided Self-Distillation (GGSD), to learn superior 3D representations from 2D pre-trained models. Specifically, we first design a geometry guided distillation module to distill knowledge from 2D models, and then leverage the 3D geometric priors to alleviate the inherent noise in 2D models and enhance the representation learning process. Due to the advantages of 3D representation, the performance of the distilled 3D student model can significantly surpass that of the 2D teacher model. This motivates us to further leverage the representation advantages of 3D data through self-distillation. As a result, our proposed GGSD approach outperforms the existing open vocabulary 3D scene understanding methods by a large margin, as demonstrated by our experiments on both indoor and outdoor benchmark datasets.