AI Paper Atlas

Research map / Detection and segmentation / Weakly supervised detection and segmentation

Calibrate Before Adapt: Training-Free Pseudo-Label Calibration for Semi-Supervised Cross-Domain Few-Shot Detection

Xin Yang, Fei Zhou, Wei Wei, Lei Zhang

ECCV 2026 · Detection and segmentation › Weakly supervised detection and segmentation · approach: Pseudo-labeling for semi-supervised detection

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 prototype-based semantic calibrator (ablation)MM-Grounding-DINO baseline · DIOR 1-shotmAP: 18.6 -> 20.6Improved
add prototype-based semantic calibrator (ablation)MM-Grounding-DINO baseline · DIOR 5-shotmAP: 29.6 -> 32.1Improved
add prototype-based semantic calibrator (ablation)MM-Grounding-DINO baseline · DIOR 10-shotmAP: 35.7 -> 36.0Improved
add prototype-based semantic calibrator (ablation)MM-Grounding-DINO baseline · NEU-DET 1-shotmAP: 13.9 -> 16.5Improved
add prototype-based semantic calibrator (ablation)MM-Grounding-DINO baseline · NEU-DET 5-shotmAP: 22.2 -> 23.5Improved
add prototype-based semantic calibrator (ablation)MM-Grounding-DINO baseline · NEU-DET 10-shotmAP: 24.3 -> 25.6Improved
add visual self-prompting with SAM on top of calibrator (ablation)calibrated pseudo-labels · DIOR 1-shotmAP: 20.6 -> 23.7Improved
add visual self-prompting with SAM on top of calibrator (ablation)calibrated pseudo-labels · DIOR 5-shotmAP: 32.1 -> 32.5Improved
add visual self-prompting with SAM on top of calibrator (ablation)calibrated pseudo-labels · NEU-DET 5-shotmAP: 23.5 -> 23.8Improved
pseudo-label filtering with threshold 0 instead of calibrationMM-Grounding-DINO · six-dataset average 1-shotmAP: 31.2 -> 19.7Hurt
use our calibration framework instead of threshold filtering thr=0.9MM-Grounding-DINO pseudo-labels · six-dataset average 1-shotmAP: 32.8 -> 35.9Improved
use visual prompts instead of text prompts for SAM3SAM3 · six-dataset average 1-shotmAP: 28.8 -> 32.1Improved

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