AI Paper Atlas

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

Hypercorrelation Squeeze for Few-Shot Segmentation

Juhong Min, Dahyun Kang, Minsu Cho

ICCV 2021 · Detection and segmentation › Open-set and few-shot detection · approach: Few-shot 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
add support feature masking (ablation)HSNet (ResNet101) · PASCAL-5i 1-shotmIoU: 61.5 -> 66.2Improved
add support feature masking (ablation)HSNet (ResNet101) · PASCAL-5i 5-shotmIoU: 63.7 -> 70.4Improved
add support feature masking (ablation)HSNet (ResNet101) · PASCAL-5i 1-shot FB-IoUFB-IoU: 72.5 -> 77.6Improved
add support feature masking (ablation)HSNet (ResNet101) · PASCAL-5i 5-shot FB-IoUFB-IoU: 73.8 -> 80.6Improved
switch backbone from ResNet50 to ResNet101HSNet · PASCAL-5i 1-shotmIoU: 64.0 -> 66.2Improved
switch backbone from ResNet50 to ResNet101HSNet · COCO-20i 1-shotmIoU: 39.2 -> 41.2Improved
increase support shots from 1 to 5HSNet (ResNet50) · COCO-20imIoU: 39.2 -> 46.9Improved
increase support shots from 1 to 5HSNet (ResNet50) · PASCAL-5imIoU: 64.0 -> 69.5Improved
compare against PFENet baselineHSNet vs PFENet (ResNet101) · COCO-20i 1-shotmIoU: 38.5 -> 41.2Improved

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