HyperAIHyperAI超神经
首页资讯最新论文教程数据集百科SOTALLM 模型天梯GPU 天梯顶会
全站搜索
关于
HyperAIHyperAI超神经

Self Supervised Image Classification On 1

评估指标

Number of Params
Top 1 Accuracy

评测结果

各个模型在此基准测试上的表现结果

模型名称
Number of Params
Top 1 Accuracy
Paper TitleRepository
DINOv2 (ViT-g/14, 448)1100M88.9%DINOv2: Learning Robust Visual Features without Supervision-
EsViT (Swin-B)87M83.9%Efficient Self-supervised Vision Transformers for Representation Learning-
iBOT (ViT-L/16)307M84.8%iBOT: Image BERT Pre-Training with Online Tokenizer-
A2MIM (ResNet-50 RSB-A2)-80.4%Architecture-Agnostic Masked Image Modeling -- From ViT back to CNN-
MoCo (Resnet-50)-77.0%Momentum Contrast for Unsupervised Visual Representation Learning-
iBOT(ViT-L/16)307M86.6%iBOT: Image BERT Pre-Training with Online Tokenizer-
MIRL (ViT-B-48)341M86.2%Masked Image Residual Learning for Scaling Deeper Vision Transformers-
SimMIM (SwinV2-H, 512)658M87.1%SimMIM: A Simple Framework for Masked Image Modeling-
DnC (Resnet-50)-78.2%Divide and Contrast: Self-supervised Learning from Uncurated Data-
A2MIM+ (ViT-B)-84.5%Architecture-Agnostic Masked Image Modeling -- From ViT back to CNN-
PercMAE (ViT-L, dVAE)307M88.6%Improving Visual Representation Learning through Perceptual Understanding-
A2MIM+ (ViT-S)-82.4%Architecture-Agnostic Masked Image Modeling -- From ViT back to CNN-
DINOv2 (ViT-g/14)1100M88.5%DINOv2: Learning Robust Visual Features without Supervision-
ResNet-152 (SparK pre-training)60M82.7%Designing BERT for Convolutional Networks: Sparse and Hierarchical Masked Modeling-
ConvNeXt-Base (SparK pre-training)89M84.8%Designing BERT for Convolutional Networks: Sparse and Hierarchical Masked Modeling-
iBOT (ViT-B/16)85M84.0%iBOT: Image BERT Pre-Training with Online Tokenizer-
BEiT-L (ViT)307M86.3%BEiT: BERT Pre-Training of Image Transformers-
MaskFeat (ViT-L)307M85.7%Masked Feature Prediction for Self-Supervised Visual Pre-Training-
A2MIM (ViT-B)-84.2%Architecture-Agnostic Masked Image Modeling -- From ViT back to CNN-
SwAV (ResNeXt-101-32x16d)193M82.0%Unsupervised Learning of Visual Features by Contrasting Cluster Assignments-
0 of 65 row(s) selected.
Self Supervised Image Classification On 1 | SOTA | HyperAI超神经