AI Paper Atlas

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

Attentive Semantic Alignment with Offset-Aware Correlation Kernels

Paul Hongsuck Seo, Jongmin Lee, Deunsol Jung, Bohyung Han, Minsu Cho

ECCV 2018 · Detection and segmentation › Open-set and few-shot detection · approach: Open-vocabulary 3D and world perception · 69 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
use proposed attentive alignment instead of GeoCNNProposed · Taniai (average)flow accuracy: 0.67 -> 0.70Improved
use proposed model instead of GeoCNN on FG3DCarProposed · Taniai FG3DCarflow accuracy: 0.85 -> 0.87Improved
use proposed model instead of GeoCNN on PASCAL subsetProposed · Taniai PASCALflow accuracy: 0.53 -> 0.55Improved
use proposed model instead of GeoCNN on Caltech-101 IoUProposed · Caltech-101IoU: 0.56 -> 0.57Improved
use proposed model instead of GeoCNN on Caltech-101 LT-ACCProposed · Caltech-101LT-ACC: 0.79 -> 0.80Improved
use proposed model instead of GeoCNN on Caltech-101 localization errorProposed · Caltech-101LOC-ERR: 0.28 -> 0.28No clear effect
combine TPS with affine instead of affine only (ImageNet objects) (ablation)Proposed · Caltech-101 (ImageNet objects)IoU: 0.54 -> 0.58Improved
combine affine with TPS instead of TPS only (ImageNet objects) (ablation)Proposed · Caltech-101 (ImageNet objects)IoU: 0.56 -> 0.58Improved
combine TPS with affine instead of affine only (other objects) (ablation)Proposed · Caltech-101 (Other objects)IoU: 0.52 -> 0.56Improved
test on objects unseen in ImageNet pretraining versus seen objects (ablation)Proposed · Caltech-101IoU: 0.58 -> 0.56No clear effect
use correlation-based alignment on cross-class matching versus best prior modelProposed · PASCAL Parts cross-classwIoU: 37.5 -> 35.0No 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