AI Paper Atlas

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

Zero-Shot Learning Using Synthesised Unseen Visual Data with Diffusion Regularisation

Yang Long, Li Liu, Fumin Shen, Ling Shao, Xuelong Li

TPAMI 2017 · Label-efficient and robust learning › Zero-/few-shot and meta-learning · approach: Generative feature synthesis for zero-shot · 105 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 graph regularisation to linear regression (ablation)linear regression synthesis · CUB unseen (CA, NN)accuracy: 27.28 -> 38.82Improved
add diffusion regularisation to linear regression (ablation)linear regression synthesis · aPY unseen (CA, NN)accuracy: 17.96 -> 30.28Improved
add diffusion regularisation to graph-only (ablation)GR-only · CUB unseen (CA, NN)accuracy: 38.82 -> 44.90Improved
add graph regularisation to diffusion-only (ablation)DR-only · aPY unseen (CA, NN)accuracy: 30.28 -> 42.25Improved
add both regularisations to linear regression (ablation)linear regression synthesis · SUN unseen (CA, NN)accuracy: 63.00 -> 80.50Improved
use SVM on synthesised samples instead of NNOurs sample-based · aPY unseenaccuracy: 38.49 -> 53.21Improved
use SVM on synthesised samples instead of NNOurs sample-based · CUB unseenaccuracy: 39.82 -> 45.72Improved
use mean of synthesised features instead of class-level attribute averageOurs prototype · CUB / SUN / aPY unseenaccuracy: CUB 44.90 -> 44.90; SUN 80.50 -> 78.50; aPY 42.25 -> 41.97No clear effect
remove diffusion regularisation: variance of synthesised features (ablation)UVDS · SUN/aPY/CUB synthesised datavariance balanceImproved
use GAN to synthesise unseen features instead of regressionGAN · AwA (VGG-19, CA, PC)per-class accuracy: 82.12 -> 62.40Hurt
use UVDS instead of baselines in GZSL on CUBSynC · CUB GZSL U-Taccuracy: 13.2 -> 23.8Improved
use UVDS on large-scale ImageNet 2-hopsSYNC · ImageNet 2Htop-1 accuracy: 9.26 -> 10.15Improved

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