AI Paper Atlas

Research map / Vision-language and multimodal / Human and embodied modeling

Open-Text Aerial Detection: A Unified Framework For Aerial Visual Grounding And Detection

Guoting Wei, Xia Yuan, Yang Zhou, Haizhao Jing, Yu Liu, Xianbiao Qi, Chunxia Zhao, Haokui Zhang, Rong Xiao

ICML 2026 · Vision-language and multimodal › Human and embodied modeling · approach: Visual grounding transformers · 0 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 attribute-level dense alignment (ablation)OTA-Det-M · AerialVGAttr-Align tau=0.5: 14.36 -> 44.23Improved
add attribute-level dense alignment (ablation)OTA-Det-M · AerialVGAttr-Align tau=0.7: 0.47 -> 21.96Improved
add attribute-level dense alignment (ablation)OTA-Det-M · AerialVGAcc@0.5: 54.67 -> 54.94No clear effect
add decoupled multi-granular head (ablation)OTA-Det-M · AerialVGAttr-Align tau=0.5: 39.55 -> 44.23Improved
add decoupled multi-granular head (ablation)OTA-Det-M · AerialVGAttr-Align tau=0.7: 14.88 -> 21.96Improved
add decoupled multi-granular head (ablation)OTA-Det-M · AerialVGAcc@0.5: 54.16 -> 54.94Improved
add image-level annotation aggregation (ablation)OTA-Det-M · AerialVGAttr-Align tau=0.5: 43.21 -> 44.23Improved
add image-level annotation aggregation (ablation)OTA-Det-M · AerialVGAttr-Align tau=0.7: 22.74 -> 21.96Hurt
add image-level annotation aggregation (ablation)OTA-Det-M · AerialVGAcc@0.5: 54.56 -> 54.94Improved
train LAE-DINO jointly on OTA-Mix with Naive ReformulationLAE-DINO · DOTAmAP: 46.8 -> 46.2Hurt
replace LAE-DINO with OTA-Det-M under task-specific LAE-1M trainingOVAD · LAE-80CmAP: +11.0 mAPImproved
replace AerialVG method with OTA-Det-MRSVG · AerialVGAcc@0.5: 50.0 -> 54.9Improved

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