AI Paper Atlas

Research map / Detection and segmentation / Open-set and few-shot detection

OVI-MAP: Open-Vocabulary Instance-Semantic Mapping

Zilong Deng, Federico Tombari, Marc Pollefeys, Johanna Wald, Daniel Barath

CVPR 2026 · Detection and segmentation › Open-set and few-shot detection · approach: Open-vocabulary semantic segmentation

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 object-centric view coverage instead of pixel-counting view selection (ablation)OVI-MAP · Replicaaverage VLM queries per instance: 18.7 -> 8.6Improved
use view coverage instead of pixel counting (ablation)OVI-MAP · ReplicaAP25: 33.2 -> 34.5Improved
use view coverage instead of pixel counting (ablation)OVI-MAP · ScanNetAP25: 24.7 -> 23.4No clear effect
use view coverage instead of random 8 views (ablation)OVI-MAP · ReplicamIoU: 23.8 -> 26.5Improved
use CropFormer instead of SAM2 for 2D entity segmentation (ablation)OVI-MAP · Replicasemantic mIoU: 20.2 -> 26.9Improved
use ground-truth 2D masks instead of CropFormer (ablation)OVI-MAP · Replicainstance AP50: 50.8 -> 65.7Improved
weight multi-view features by visible pixel count instead of plain averaging (ablation)OVI-MAP · ReplicaAP50: 19.8 -> 22.0Improved
use feature clustering instead of averaging (ablation)OVI-MAP · ReplicamIoU: 26.5 -> 25.1Hurt
use SigLIP-large instead of CLIP ViT-L as VLM backbone (ablation)OVI-MAP · ReplicamIoU: 15.4 -> 26.9Improved
use larger SigLIP2 so400m instead of SigLIP-large (ablation)OVI-MAP · ReplicamIoU: 26.9 -> 26.4No clear effect
reduce frames processed to meet 30 FPSOVO-SLAM · ReplicamIoU: 24.9 -> 21.8Hurt
reduce frames processed to meet 30 FPSOVI-MAP · ReplicamIoU: 26.5 -> 27.0No clear effect

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