GaussianVLM embeds language features per Gaussian splat, sparsifies them by task and location, and reports state-of-the-art results on embodied 3D reasoning benchmarks.
Scanqa: 3d question answering for spatial scene understanding
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
GaussianVLM: Scene-centric 3D Vision-Language Models using Language-aligned Gaussian Splats for Embodied Reasoning and Beyond
GaussianVLM embeds language features per Gaussian splat, sparsifies them by task and location, and reports state-of-the-art results on embodied 3D reasoning benchmarks.