AI Paper Atlas

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

VISO: Accelerating In-orbit Object Detection with Language-Guided Mask Learning and Sparse Inference

Meiqi Wang, Han Qiu

ICCV 2025 · Detection and segmentation › Open-set and few-shot detection · approach: Open-vocabulary object detection · 1 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
add language-guided attention module (ablation)VISO-L · zero-shot MAR20, HRSC2016, VEDAIAP: 0.4 -> 1.9Improved
add language-guided attention module (ablation)VISO-L · VEDAI (zero-shot)AP: 48.3 -> 50.2Improved
fine-tune CLIP text encoder instead of freezing (ablation)VISO-L · VEDAI (zero-shot)AP: 45.0 -> 50.2Improved
fine-tune CLIP text encoder instead of freezing (ablation)VISO-L · MAR20 (zero-shot)AP: 89.9 -> 90.0Improved
add visual grounding and more object detection pre-training data (ablation)VISO-L · zero-shot benchmarksAP: 0.3 -> 63.9Improved
add visual grounding data to OD pre-training (ablation)VISO-L · HRSC2016 (zero-shot)AP: 68.9 -> 70.7Improved
apply sparse inference (ablation)VISO-S · MAR20 (zero-shot)AP: 87.6 -> 87.4No clear effect
apply sparse inference (ablation)VISO-L · DOTAv2 testFLOPs: 229 -> 128Improved
apply sparse inference at tuned threshold (ablation)VISO-S · DOTAv2 testAP: 56.8 -> 50.6Hurt
apply sparse inference (ablation)VISO-L · DOTAv2 testAP: 63.4 -> 61.1Hurt
apply sparse inference (ablation)VISO-S · DIOR-RSVGPr@0.5: 74.4 -> 70.1Hurt
apply sparse inference (ablation)VISO-L · VEDAI (zero-shot)AP: 50.2 -> 49.9No clear effect

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