AI Paper Atlas

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

Variational Feature Disentangling for Fine-Grained Few-Shot Classification

Jingyi Xu, Hieu Le, Mingzhen Huang, ShahRukh Athar, Dimitris Samaras

ICCV 2021 · Label-efficient and robust learning › Zero-/few-shot and meta-learning · approach: Embedding adaptation for few-shot classification · 86 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 augmented features from the instance-conditioned posterior with 1NN (ablation)VFD · CUB 5-way 1NNaccuracy: 74.95 -> 75.46Improved
add augmented features from Delta-encoder with 1NNDelta-encoder · CUB 5-way 1NNaccuracy: 69.37 -> 67.31Hurt
add Delta-encoder augmentation at 20-way with 1NNDelta-encoder · CUB 20-way 1NNaccuracy: 48.10 -> 38.84Hurt
add our augmentation at 10-way with 1NN (ablation)VFD · CUB 10-way 1NNaccuracy: 62.05 -> 63.17Improved
sample variance features from instance-conditioned posterior instead of N(0,I) (ablation)VFD · CUB and NAB 5-way 1-shotaccuracyImproved
use isotropic Gaussian prior instead of Gaussian mixture prior (ablation)VFD · CUB 5-way 1-shotaccuracy: 75.16 -> 79.12Improved
use variational inference instead of covariance matrix estimate (ablation)VFD · CUB 5-way 1-shotaccuracy: 75.28 -> 79.12Improved
add feature disentanglement of intra-class variance (ablation)VFD · CUB 5-way 1-shotaccuracy: 73.40 -> 79.12Improved
add feature disentanglement of intra-class variance (ablation)VFD · NAB 5-way 1-shotaccuracy: 81.83 -> 88.62Improved
use VFD instead of MetaOptNetVFD · CUB 5-way 5-shot (ResNet12)accuracy: 87.09 -> 91.48Improved
use VFD instead of Delta-encoder as augmentationVFD · Stanford Dogs 5-way 5-shot (ResNet12)accuracy: 78.60 -> 88.00Improved
use VFD instead of MattMLVFD · CUB 5-way 1-shot (Conv4)accuracy: 66.29 -> 68.42Improved

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