ModelMAE / Masked Autoencoders
MAE
Masked Autoencoders，一种基于掩码重建的自监督视觉学习方法，通过遮盖图像部分区域并预测被遮盖内容来训练模型
Timeline (last 90 days)
Aug 4
The choice of pretraining (ImageNet-1K vs ImageNet-21K, supervised vs self-supervised like MAE/DINO) significantly impacts downstream performance, potentially causing several percentage points of difference
Unverified80%