ML Research Wiki / Benchmarks / Long-tail Learning / EGTEA

EGTEA

Long-tail Learning Benchmark

Performance Over Time

📊 Showing 3 results | 📏 Metric: Average Precision

Top Performing Models

Rank Model Paper Average Precision Date Code
1 CDB-loss (3D- ResNeXt101) Class-Wise Difficulty-Balanced Loss for Solving Class-Imbalance 66.24 2020-10-05 📦 hitachi-rd-cv/CDB-loss
2 CB Loss Class-Balanced Loss Based on Effective Number of Samples 63.26 2019-01-16 📦 vandit15/Class-balanced-loss-pytorch 📦 richardaecn/class-balanced-loss 📦 frgfm/Holocron
3 Focal loss (3D- ResNeXt101) Focal Loss for Dense Object Detection 59.17 2017-08-07 📦 tensorflow/models 📦 facebookresearch/detectron2 📦 open-mmlab/mmdetection

All Papers (3)

hitachi-rd-cv/CDB-loss
vandit15/Class-balanced-loss-pytorch richardaecn/class-balanced-loss

Focal Loss for Dense Object Detection

2017
Focal loss (3D- ResNeXt101)
tensorflow/models facebookresearch/detectron2