AI Paper Atlas

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

Reference Twice: A Simple and Unified Baseline for Few-Shot Instance Segmentation

Yue Han, Jiangning Zhang, Yabiao Wang, Chengjie Wang, Yong Liu, Lu Qi, Ming–Hsuan Yang, Xiangtai Li

TPAMI 2024 · Detection and segmentation › Open-set and few-shot detection · approach: Few-shot segmentation · 33 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 query-based second reference module (ablation)Mask2Former baseline · COCO minival K=10nAP: 14.7 -> 16.3Improved
add image-based first reference module on top of second reference (ablation)RefT · COCO minival K=10nAP: +1.1Improved
second reference reduces base-class bias (ablation)Mask2Former · COCO minival K=10bAP: +1.6 nAP, -0.5 bAPMixed
first reference helps both novel and base classes (ablation)Mask2Former · COCO minival K=10nAP and bAP: +0.9 nAP, +0.3 bAPImproved
replace cross-attention with adaptive convolution in second reference (ablation)RefT · COCO gFSIS K=10nAP: 17.4 -> 16.8Hurt
add auxiliary classification loss in second reference (ablation)RefT · COCO gFSIS K=10nAP: 17.4 -> 17.5No clear effect
fine-tune CAL during novel fine-tuning (vs frozen) (ablation)RefT · COCO minival K=10nAP: 2.6 -> 17.4Improved
use larger-scale support features (Res3 vs Res5) in first reference (ablation)RefT · COCO minival K=10nAP: 16.4 -> 17.2Improved
use query-level instead of feature-level enhancement in second reference (ablation)RefT · COCO minival K=10nAP: +1.1Improved
add Class-Enhanced BKD Loss instead of L2 weight regularization (ablation)RefT · COCO iFSIS K=10bAP: 29.4 -> 33.2Improved
use BKD loss instead of weight regularization (ablation)RefT · COCO iFSIS K=10nAP: 11.5 -> 10.1Hurt
use Bayesian logit classifier instead of cosine classifier (ablation)RefT · COCO iFSIS K=10nAP: +1.6 nAP, +0.8 bAPImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 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