Action Segmentation On Jigsaws
評価指標
Accuracy
Edit Distance
F1@10
F1@25
F1@50
評価結果
このベンチマークにおける各モデルのパフォーマンス結果
モデル名 | Accuracy | Edit Distance | F1@10 | F1@25 | F1@50 | Paper Title | Repository |
|---|---|---|---|---|---|---|---|
| RL (full) | 81.43 | 87.96 | 92.0 | 90.5 | 82.2 | Deep Reinforcement Learning for Surgical Gesture Segmentation and Classification | |
| ST-CNN+Seg | 74.22 | 66.56 | - | - | - | Segmental Spatiotemporal CNNs for Fine-grained Action Segmentation | - |
| TricorNet | 82.9 | 86.8 | - | - | - | TricorNet: A Hybrid Temporal Convolutional and Recurrent Network for Video Action Segmentation | - |
| RL+Tree | 81.67 | 88.53 | 92.68 | 90.99 | 83.15 | Automatic Gesture Recognition in Robot-assisted Surgery with Reinforcement Learning and Tree Search | - |
| SDL+SC-CRF | - | 86.21 | - | - | - | End-to-End Fine-Grained Action Segmentation and Recognition Using Conditional Random Field Models and Discriminative Sparse Coding | - |
| TCN | 81.4 | 83.1 | - | - | - | Temporal Convolutional Networks: A Unified Approach to Action Segmentation | |
| MRG-Net | 87.9±4.2 | 89.3±5.2 | - | - | - | Relational Graph Learning on Visual and Kinematics Embeddings for Accurate Gesture Recognition in Robotic Surgery | - |
0 of 7 row(s) selected.