AI Paper Atlas

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

Disentangling the Benefits of Self-Supervised Learning to Deployment-Driven Downstream Tasks of Satellite Images (Student Abstract)

Zhuo Deng, Yibing Wei, Mingye Zhu, Xueliang Wang, Junchi Zhou, Zhicheng Yang, Hang Zhou, Zhenjie Cao, Lan Ma, Mei Han, Jui-Hsin Lai

AAAI 2023 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Data-efficient and robust self-supervision · 2 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
Use SimCLR pretraining on BEN3 instead of ImageNet RGB pretrainingResNet50 · NaSC-TG2, Train/Validation ratios 2:8, 5:5, and 8:2Top-1 accuracy: 92.85 -> 92.25; 93.64 -> 93.37; 94.05 -> 94.45Mixed
Use four spectral channels (BEN4, RGBN) rather than three (BEN3, RGB) for SimCLR pretrainingResNet50 on NaSC-TG2 · NaSC-TG2, Train/Validation ratios 2:8, 5:5, and 8:2Top-1 accuracy: 92.25 -> 92.90; 93.37 -> 95.59; 94.45 -> 97.13Improved
Use MAE+ViT-B with BEN4NaSC-TG2 scene classification · NaSC-TG2, Train/Validation ratios 2:8, 5:5, and 8:2Top-1 accuracy: 93.28; 96.11; 98.30Improved
Use SimCLR-pretrained ResNet50 rather than ResNet50 without SSLPotsdam segmentation · PotsdammIoU: 80.80 -> 82.76Improved
Use SimCLR-pretrained ResNet50 rather than ResNet50 without SSLPotsdam segmentation · PotsdamOA: 89.80 -> 91.01Improved
Add MAE pretraining on BEN4ViT-B+UperNet on Potsdam · PotsdammIoU: 76.57 -> 78.35Improved
Add MAE pretraining on BEN4ViT-B+UperNet on Potsdam · PotsdamOA: 87.33 -> 88.41Improved
Use vanilla ViT-B rather than the CNN-based ResNet50 modelPotsdam segmentation without SSL · PotsdammIoU: 80.80 -> 76.57Hurt

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