SceneSplat++: A Large Dataset and Comprehensive Benchmark for Language Gaussian Splatting

Mengjiao Ma, Qi Ma, Yue Li, Jiahuan Cheng, Runyi Yang, Bin Ren, Nikola Popovic, Mingqiang Wei, Nicu Sebe, Ender Konukoglu, Luc Van Gool, Theo Gevers, Martin R. Oswald, Danda Pani Paudel

Sofia University· Nanjing University of Aeronautics and Astronautics· ETH· University of Amsterdam· Johns Hopkins University· University of Pisa· University of Trento

open vocabulary scene understanding benchmark 3DGS scene dataset

⋅ NeurIPS ⋅ Project Page ⋅OpenReview

Abstract

3D Gaussian Splatting (3DGS) serves as a highly performant and efficient encoding of scene geometry, appearance, and semantics. Moreover, grounding language in 3D scenes has proven to be an effective strategy for 3D scene understanding.

Current Language Gaussian Splatting line of work fall into three main groups:

per-scene optimization-based,
per-scene optimization-free, and
generalizable approach.

However, most of them are evaluated only on rendered 2D views of a handful of scenes and viewpoints close to the training views, limiting ability and insight into holistic 3D understanding.

To address this gap, we propose the first large-scale benchmark that systematically assesses these three groups of methods directly in 3D space, evaluating on 1060 scenes across three indoor datasets and one outdoor dataset. Benchmark results demonstrate a clear advantage of the generalizable paradigm, particularly in relaxing the scene-specific limitation, enabling fast feed-forward inference on novel scenes, and achieving superior segmentation performance.

We further introduce SceneSplat-49K -- a carefully curated 3DGS dataset comprising of around 49K diverse indoor and outdoor scenes trained from multiple sources, with which we demonstrate generalizable approach could harness strong data priors. Our codes, benchmark, and datasets are available.