AI Paper Atlas

Research map / 3D vision / Gaussian splatting

Open-Set Semantic Gaussian Splatting SLAM with Expandable Representation

Yucheng Yan, Chen Liang, Wenguan Wang, Yi Yang

ICLR 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 semantic feature pool FP (ablation)SplaTAM Ds=3 · Replica Room 0mIoU: 72.56 -> 80.15Improved
add Intra-Inter Semantic Consistency Objective LCO (ablation)SplaTAM + FP · Replica Room 0mIoU: 80.15 -> 84.23Improved
add Semantic Stability Guidance MSG (ablation)SplaTAM + FP + LCO · Replica Room 0mIoU: 84.23 -> 87.29Improved
add semantic components to SplaTAMSplaTAM · ReplicaPSNR: 34.11 -> 37.61Improved
add semantic components to SplaTAMSplaTAM · ReplicaATE RMSE (cm): 0.36 -> 0.29Improved
add semantic components to LoopSplatLoopSplat · ReplicaPSNR: 36.63 -> 37.95Improved
add semantic components to LoopSplatLoopSplat · ReplicaDepth L1 (cm): 0.51 -> 0.30Improved
use expandable pool instead of fixed-size explicit pool (ablation)semantic pool · Replica Room 0mIoU: 86.94 -> 96.93Improved
use expandable pool instead of compressive pool (ablation)semantic pool · Replica Room 0mIoU: 88.62 -> 96.93Improved
increase aggregation neighbors m from 1 to 5 (ablation)semantic pool · Replica Room 0mIoU: 58.12 -> 96.93Improved
raise novelty threshold from 0.6 to 0.8 (ablation)semantic pool · Replica Room 0mIoU: 81.67 -> 96.93Improved
raise semantic feature dimension Ds from 3 to 32 (ablation)semantic pool · Replica Room 0mIoU: 87.29 -> 96.93Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 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