Jul 18, 2024 � In this paper, we propose an effective approach, namely Geometry Guided Self-Distillation (GGSD), to learn superior 3D representations from 2D pre-trained�...
In this paper, we propose an effective approach, namely Geometry Guided Self-Distillation (GGSD), to learn superior 3D representations from 2D pre-trained�...
Official PyTorch codes for "Open Vocabulary 3D Scene Understanding via Geometry Guided Self-Distillation", ECCV2024
Jul 18, 2024 � This paper presents a novel approach for open vocabulary 3D scene understanding that leverages 3D geometric information to guide the learning of�...
In this paper, we propose an effective approach, namely Geometry Guided Self-Distillation (GGSD), to learn superior 3D representations from 2D pre-trained�...
People also ask
What is 3D scene understanding?
What is self distillation in deep learning?
Open-vocabulary scene understanding aims to localize and recognize unseen categories beyond the annotated la- bel space. The recent breakthrough of 2D open-�...
We propose OpenScene, a zero-shot method for extract- ing 3D dense features from an open vocabulary embed- ding space using multi-view fusion and 3D convolution�...
Open Vocabulary 3D Scene Understanding via Geometry Guided Self-Distillation ... open vocabulary capabilities to 3D models through knowledge distillation.
Oct 10, 2024 � Open Vocabulary 3D Scene Understanding via Geometry Guided Self-Distillation. Pengfei Wang, Yuxi Wang, Shuai Li, Zhaoxiang Zhang, Zhen Lei�...
Extensive experiments show that the method even outperforms fully supervised models trained with segmentation annotations in certain scenes.