Welcome to the DeepLabCut Benchmark!

This page hosts the official DeepLabCut Benchmark.

Leaderboard

Click on columns to sort in ascending or descending order. Press Ctrl while selecting columns to sort by multiple columns. Hover over a column name to get additional explanations for the metric.

Model Code Dataset Pose - RMSE Pose - mAP
- EfficientNet B7_s4 (30k) https://github.com/DeepLabCut/DeepLabCut TRI-MOUSE 4.93 0.943403
- DLCRNet_ms4 (30k) https://github.com/DeepLabCut/DeepLabCut TRI-MOUSE 3.55 0.931542
- ResNet50 (30k) https://github.com/DeepLabCut/DeepLabCut TRI-MOUSE 3.97 0.927728
- EfficientNet B7_s4 (30k) https://github.com/DeepLabCut/DeepLabCut PARENTING 8.9 0.639412
- DLCRNet_ms4 (30k) https://github.com/DeepLabCut/DeepLabCut PARENTING 9.85 0.62907
- EfficientNet B7 (30k) https://github.com/DeepLabCut/DeepLabCut PARENTING 10.24 0.627089
- EfficientNet B7_s4 (30k) https://github.com/DeepLabCut/DeepLabCut FISH 6.85 0.753309
- DLCRNet_ms4 (30k) https://github.com/DeepLabCut/DeepLabCut FISH 7.15 0.754255
- ResNet50_s4 (30k) https://github.com/DeepLabCut/DeepLabCut FISH 7.03 0.741195
- DLCRNet_ms4 (200k) https://github.com/DeepLabCut/DeepLabCut MARMOSETS 7.56 0.894922
- EfficientNet B7 (200k) https://github.com/DeepLabCut/DeepLabCut MARMOSETS 7.27 0.888887
- EfficientNet B7_s4 (200k) https://github.com/DeepLabCut/DeepLabCut MARMOSETS 6.83 0.893077
- DLCRNet (200k) https://github.com/DeepLabCut/DeepLabCut MARMOSETS 8.1 0.883469