AI Paper Atlas

Research map / Label-efficient and robust learning / Self-supervised and contrastive

Replacing Labeled Real-Image Datasets With Auto-Generated Contours

Hirokatsu Kataoka, Ryo Hayamizu, Ryosuke Yamada, Kodai Nakashima, Sora Takashima, Xinyu Zhang, Edgar Josafat Martinez-Noriega, Nakamasa Inoue, Rio Yokota

CVPR 2022 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Pretext-task visual pre-training · 32 citations (OpenAlex)

Check 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
pre-train on ExFractalDB-21k instead of ImageNet-21kViT-Base · ImageNet-1ktop-1 accuracy: 81.8 -> 82.7Improved
pre-train on RCDB-21k instead of ImageNet-21kViT-Base · ImageNet-1ktop-1 accuracy: 81.8 -> 82.4Improved
pre-train on ExFractalDB-21k instead of ImageNet-21kSwin Transformer · COCO detectionAP: 48.8 -> 48.0Hurt
pre-train on ExFractalDB-21k instead of ImageNet-21kSwin Transformer · COCO instance segmentationAP: 43.6 -> 42.8Hurt
use ExFractalDB-1k pre-training instead of ImageNet-1k supervised pre-trainingfine-tuned models · Stanford Carsaccuracy: 89.9 -> 90.3Improved
use ExFractalDB-1k* pre-training instead of ImageNet-1k supervised pre-trainingfine-tuned models · Flowersaccuracy: 99.4 -> 99.6Improved
increase the number of varied fractal parameters from three to six (ablation)FractalDB pre-training · CIFAR-10accuracy: 96.8 -> 97.0Improved
use contour vertex counts of 203–502 instead of 3–102 (ablation)RCDB pre-training · CIFAR-10accuracy: 95.5 -> 71.3; 95.5 -> 59.4; 95.5 -> 40.1Hurt
add 1k background-colored lines to break object contours (ablation)RCDB pre-training · CIFAR-10accuracy: 95.5 -> 46.2Hurt
use a line-counting task with consistent labels instead of randomly permuted labels (ablation)LineDB pre-training · CIFAR-10accuracy: 13.5 -> 95.6Improved
increase the number of line-counting categories from 512 to 1k (ablation)LineDB pre-training · CIFAR-10accuracyHurt

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