atlas

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

Disentangling Visual Embeddings for Attributes and Objects

CVPR 2022 · Detection and segmentation › Open-set and few-shot detection · approach: Zero-shot segmentation and detection · 82 citations (OpenAlex)

Check your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's title only (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 GloVe instead of Fasttext embeddings (ablation)OADis · MIT-StatesAUC@1: 5.3 -> 5.9Helped
Combine GloVe with Fasttext embeddings instead of using GloVe alone (ablation)OADis · MIT-StatesAUC@1: 5.9 -> 5.5Hurt
Set lambda to 0.1 instead of 10 (ablation)OADis · MIT-StatesAUC@1: 5.9 -> 5.7Hurt
Set delta to 0.01 instead of 0.05 (ablation)OADis · MIT-StatesAUC@1: 5.9 -> 4.8Hurt
Set delta to 0.1 instead of 0.05 (ablation)OADis · MIT-StatesAUC@1: 5.9 -> 5.2Hurt
Set alpha1/alpha2 to 0.5 and alpha3/alpha4 to 0.1 rather than 0.05 (ablation)OADis · MIT-StatesAUC@1: 5.9 -> 5.3Hurt
Use an Object-Conditioned network rather than a Linear or MLP network (ablation)OADis · MIT-StatesAUC@1Helped

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