AI Paper Atlas

Research map / Image generation / Text-to-image diffusion

IS SYNTHETIC DATA FROM GENERATIVE MODELS READY FOR IMAGE RECOGNITION?

Ruifei He, Shuyang Sun, Xin Yu, Chuhui Xue, Wenqing Zhang, Philip Torr, Song Bai, XIAOJUAN QI

ICLR 2023 · Image generation › Text-to-image diffusion · approach: Personalized and region-controlled generation · 54 citations (OpenAlex)

Read the paper

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 synthetic data (LE+CF+SCE) to classifier tuning in zero-shot settingCLIP-RN50 · 17 zero-shot datasetstop-1 accuracy (average): +4.31Improved
add synthetic data (LE+CF+SCE) to classifier tuning in zero-shot settingCLIP-ViT-B/16 · 17 zero-shot datasetstop-1 accuracy (average): +2.90Improved
add synthetic data to zero-shot classifier tuningCLIP-RN50 · EuroSAT zero-shottop-1 accuracy: +17.86Improved
add language enhancement (LE) over basic label-name prompts (ablation)zero-shot synthetic data · EuroSAT and CIFAR-100 zero-shottop-1 accuracy: +6.71 on EuroSAT (CE); +0.66 on CIFAR-100Improved
add language enhancement (LE) over basic label-name prompts (ablation)zero-shot synthetic data · CIFAR-10 zero-shottop-1 accuracy: -0.68 (CE, LE vs B)Hurt
add CLIP-based filtering (CF) to LE (ablation)zero-shot synthetic data · CIFAR-10, CIFAR-100, EuroSAT zero-shottop-1 accuracy: CIFAR-10 77.20 to 80.01 (CE)Improved
soft cross-entropy (SCE) instead of CE loss (ablation)zero-shot synthetic data · CIFAR-10, CIFAR-100, EuroSAT zero-shottop-1 accuracy: +0.05 to +1.14 over CE (LE+CF: CIFAR-100 44.55 to 45.69)Improved
train ResNet-50 from scratch on synthetic data onlyResNet-50 on CIFAR-100 · CIFAR-100 from scratchtop-1 accuracyHurt
tune more encoder parameters with synthetic data instead of classifier only (ablation)CLIP-RN50 zero-shot on EuroSAT · EuroSAT zero-shottop-1 accuracy: 55.37 (0.02%) down to 52.09 (92.19%)Hurt
add synthetic data to few-shot training (basic strategy) (ablation)CLIP-ViT-B/16 few-shot · EuroSAT 16-shotaccuracy: 87.1 vs 85.83 (no syn)Improved
real guidance (RG) for synthetic generation vs basic (ablation)few-shot synthetic data · EuroSAT 16-shotaccuracy: +1.37 over BImproved
real filtering (RF) of synthetic images (ablation)few-shot synthetic data · EuroSAT 16-shotaccuracy: +0.23 over BImproved

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