GAGS: Granularity-Aware Feature Distillation for Language Gaussian Splatting
This work addresses a key bottleneck in 3D open-vocabulary scene understanding for applications like robotics and AR/VR, though it is incremental as it builds on existing methods like CLIP and Gaussian splatting.
The paper tackles the problem of multiview inconsistency in distilling 2D CLIP features into 3D Gaussian splatting for open-vocabulary scene understanding, achieving significant performance and stability improvements with an inference speed 2× faster than baseline methods.
3D open-vocabulary scene understanding, which accurately perceives complex semantic properties of objects in space, has gained significant attention in recent years. In this paper, we propose GAGS, a framework that distills 2D CLIP features into 3D Gaussian splatting, enabling open-vocabulary queries for renderings on arbitrary viewpoints. The main challenge of distilling 2D features for 3D fields lies in the multiview inconsistency of extracted 2D features, which provides unstable supervision for the 3D feature field. GAGS addresses this challenge with two novel strategies. First, GAGS associates the prompt point density of SAM with the camera distances, which significantly improves the multiview consistency of segmentation results. Second, GAGS further decodes a granularity factor to guide the distillation process and this granularity factor can be learned in a unsupervised manner to only select the multiview consistent 2D features in the distillation process. Experimental results on two datasets demonstrate significant performance and stability improvements of GAGS in visual grounding and semantic segmentation, with an inference speed 2$\times$ faster than baseline methods. The code and additional results are available at https://pz0826.github.io/GAGS-Webpage/ .