AI Paper Atlas

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

UniFS: Universal Few-shot Instance Perception with Point Representations

Sheng Jin, Ruijie Yao, Lumin Xu, Wentao Liu, Chen Qian, Ji Wu, Ping Luo

ECCV 2024 · Detection and segmentation › Open-set and few-shot detection · approach: Few-shot object detection · 3 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 SAPL (1-hop) to L1 point loss (ablation)UniFS · COCO-UniFS val K=1Segmentation AP: 7.2 -> 8.4Improved
extend SAPL from 1-hop to 2-hop (ablation)UniFS · COCO-UniFS val K=1Segmentation AP: 8.4 -> 8.6Improved
extend SAPL to 3 or 4 hops (ablation)UniFS · COCO-UniFS val K=5Keypoint AP: 22.1 -> 20.9Hurt
use L1 loss instead of L2 for point regression (ablation)UniFS · COCO-UniFS val K=1Segmentation AP: 6.1 -> 7.2Improved
add segmentation and keypoint tasks to joint training (ablation)UniFS · COCO-UniFS val K=1Detection AP: 12.2 -> 12.7Improved
add keypoint task to Det.+Seg. training (ablation)UniFS · COCO-UniFS val K=5Detection AP: 17.6 -> 18.2Improved
add segmentation to detection-only training (ablation)UniFS · COCO-UniFS val K=5Detection AP: 17.9 -> 17.6No clear effect
increase segmentation points from 32 to 64 (ablation)UniFS · COCO-UniFS valDetection AP: Seg AP K=5 11.2 -> 11.4; Det AP K=5 17.6 -> 17.0Mixed
increase point decoder blocks beyond 2 (to 4 or 6) (ablation)UniFS · COCO-UniFS val K=5Detection AP: 18.2 -> 17.2Hurt
use UniFS single model instead of Mask-RCNN-ft multi-task baselinemulti-task few-shot learner · COCO-UniFS val K=5Detection AP: 6.9 -> 18.2Improved
UniFS versus DCFS on 5-shot segmentationUniFS · COCO-UniFS val K=5Segmentation AP: 13.5 -> 11.5Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 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