AI Paper Atlas

Research map / 3D vision / Gaussian splatting

SplatTalk: 3D VQA with Gaussian Splatting

Anh Thai, Songyou Peng, Kyle Genova, Leonidas Guibas, Thomas Funkhouser

ICCV 2025 · 3D vision › Gaussian splatting · approach: Compact and efficient Gaussian splatting · 28 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use entropy-adaptive Gaussian sampling instead of random sampling (ablation)SplatTalk · ScanQA (val)Avg: 26.73 -> 26.90Improved
use entropy-adaptive sampling instead of point-density sampling (ablation)SplatTalk · ScanQA (val)Avg: 26.00 -> 26.90Improved
use entropy-adaptive sampling instead of FPS (ablation)SplatTalk · ScanQA (val)Avg: 26.78 -> 26.90Improved
use entropy-adaptive sampling instead of FPS (ablation)SplatTalk · ScanQA (val)EM@1: 17.7 -> 17.1Hurt
increase visual input from 729 to 32,076 tokens (ablation)SplatTalk · ScanQA (val)EM@1: 16.2 -> 17.1Improved
increase visual input from 729 to 32,076 tokens (ablation)SplatTalk · SQA3D (test)EM@1-R: 38.0 -> 41.3Improved
increase visual input from 729 to 32,076 tokens (ablation)SplatTalk · MSR3D (test)EM@1: 7.9 -> 14.1Improved
use 3D Gaussian features instead of 2D LLaVA-OV featuresSplatTalk · ScanQA (val)CIDEr: 50.0 -> 61.7Improved
fine-tune SplatTalk on ScanQASplatTalk · ScanQA (val)CIDEr: 61.7 -> 77.1Improved
fine-tune SplatTalk on ScanQA and SQA3DSplatTalk · SQA3D (test)EM@1: 26.1 -> 47.6Improved
fine-tune SplatTalk on ScanQASplatTalk · MSR3D zero-shotOverall LLM-Match: 41.8 -> 41.5No clear effect
fine-tune SplatTalk on ScanQA and SQA3DSplatTalk · MSR3D zero-shotOverall LLM-Match: 41.8 -> 26.5Hurt

Later papers that compared with this method

2 later accepted papers report SplatTalk as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy