AI Paper Atlas

Research map / 3D vision / Gaussian splatting

ExtrinSplat: Decoupling Geometry and Semantics for Open-Vocabulary Understanding in 3D Gaussian Splatting

Jiayu Ding, Xinpeng Liu, Zhiyi Pan, Shiqiang Long, Ge Li

CVPR 2026 · 3D vision › Gaussian splatting · approach: Compact and efficient Gaussian splatting

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
add entropy filter and opacity filter together for neutral points (ablation)ExtrinSplat · LERF meanmIoU (%): 53.0 -> 54.3Improved
add entropy filter alone (ablation)ExtrinSplat · LERF meanmIoU (%): 53.0 -> 53.2No clear effect
add opacity filter alone (ablation)ExtrinSplat · LERF meanmIoU (%): 53.0 -> 52.6Hurt
use VLM-generated text hypotheses instead of single-view CLIP image feature (ablation)ExtrinSplat · LERF meanmIoU (%): 36.9 -> 54.3Improved
use VLM text hypotheses instead of averaged all-view CLIP image features (ablation)ExtrinSplat · LERF meanmIoU (%): 39.2 -> 54.3Improved
use VLM text hypotheses instead of IoU-filtered averaged CLIP image features (ablation)ExtrinSplat · LERF meanmIoU (%): 50.1 -> 54.3Improved
use stronger VLM (Gemini 2.5 Pro vs InternVL3-78B vs SenseNova 6.5 Pro) (ablation)ExtrinSplat · LERF meanmIoU (%): 47.0 -> 50.2 -> 54.3Improved
increase number of text hypotheses per instance from 1 to 5 (ablation)ExtrinSplat · LERF meanmIoU (%): 44.0 -> 54.3Improved
increase number of text hypotheses from 5 to 10 (ablation)ExtrinSplat · LERF meanmIoU (%): 54.3 -> 53.6Hurt
set opacity threshold to 0 (remove all high-entropy points) (ablation)ExtrinSplat · LERF meanmIoU (%): 54.3 -> 53.2Hurt
use extrinsic index layer instead of embedding per-Gaussian featuresExtrinSplat · LERF figurinesCLIP feature storage and VRAM: ~3GB -> ~3MB; ~8 GB peak VRAMImproved
use ExtrinSplat vs best prior training-free method on LERFExtrinSplat · LERF meanmIoU (%): 50.4 -> 54.3Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

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