SLAG: Scalable Language-Augmented Gaussian Splatting

Abstract

Language-augmented scene representations hold great promise for large-scale robotics applications such as searchand- rescue, smart cities, and mining. Many of these scenarios are time-sensitive, requiring rapid scene encoding while also being data-intensive, necessitating scalable solutions. Deploying these representations on robots with limited computational resources further adds to the challenge. To address this, we introduce SLAG, a multi-GPU framework for language-augmented Gaussian splatting that enhances the speed and scalability of embedding large scenes. Our method integrates 2D visual-language model features into 3D scenes using SAM [1] and CLIP [2]. Unlike prior approaches, SLAG eliminates the need for a loss function to compute per- Gaussian language embeddings. Instead, it derives embeddings from 3D Gaussian scene parameters via a normalized weighted average, enabling highly parallelized scene encoding. Additionally, we introduce a vector database for efficient embedding storage and retrieval. Our experiments show that SLAG achieves an 18× speedup in embedding computation on a 16-GPU setup compared to OpenGaussian [3], while preserving embedding quality on the ScanNet [4] and LERF [5] datasets. For more details, visit our project website: <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://slag-project.github.io/</uri>.