AI Paper Atlas

Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning

Relational Embedding for Few-Shot Classification

Dahyun Kang, Heeseung Kwon, Juhong Min, Minsu Cho

ICCV 2021 · Label-efficient and robust learning › Zero-/few-shot and meta-learning · approach: Fine-grained and cross-domain few-shot baselines · 299 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 self-correlational representation (SCR) (ablation)GAP baseline · miniImageNetaccuracy (5-way 1-shot): 65.38 -> 66.66Improved
add self-correlational representation (SCR) (ablation)GAP baseline · CUB-200-2011accuracy (5-way 1-shot): 77.54 -> 78.69Improved
add cross-correlational attention (CCA) (ablation)GAP baseline · miniImageNetaccuracy (5-way 1-shot): 65.38 -> 66.00Improved
add cross-correlational attention (CCA) (ablation)GAP baseline · CUB-200-2011accuracy (5-way 1-shot): 77.54 -> 78.49Improved
add both SCR and CCA (ablation)GAP baseline · miniImageNetaccuracy (5-way 1-shot): 65.38 -> 67.60Improved
add both SCR and CCA (ablation)GAP baseline · CUB-200-2011accuracy (5-way 1-shot): 77.54 -> 79.49Improved
use SCR instead of non-local self-attention (ablation)RENet · miniImageNetaccuracy (5-way 1-shot): 65.00 -> 66.66Improved
use channel-wise self-correlation instead of ordered cosine similarity (SCE) (ablation)SCR · miniImageNetaccuracy (5-way 1-shot): 63.39 -> 66.66Improved
use CCA instead of CAN co-attention (ablation)RENet · miniImageNetaccuracy (5-way 1-shot): 65.66 -> 66.00Improved
use relative instead of absolute neighborhood in self-correlation (ablation)SCR · CUB-200-2011accuracy (5-way 1-shot): 76.34 -> 78.69Improved
use relative instead of absolute neighborhood in self-correlation (ablation)SCR · miniImageNetaccuracy (5-way 1-shot): 66.41 -> 66.66Improved
use separable instead of vanilla 4D convolution in CCA (ablation)CCA · miniImageNet / CUBaccuracy (5-way 1-shot): 65.59 -> 65.90 (mini), 78.89 -> 78.49 (CUB), GPU 60.35 -> 34.97 msNo clear effect

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