Visual Object Tracking On Trackingnet
評価指標
Accuracy
Normalized Precision
Precision
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
比較表
| モデル名 | Accuracy | Normalized Precision | Precision |
|---|---|---|---|
| autoregressive-visual-tracking | 85.6 | 89.6 | 86.0 |
| siamfc-towards-robust-and-accurate-visual | 74.5 | 79.8 | 68.5 |
| dctd-deep-conditional-target-densities-for | - | 80.1 | 69.7 |
| neighbortrack-improving-single-object | 83.79 | 88.30 | - |
| joint-group-feature-selection-and | 60.9 | 71.79 | 56.57 |
| learning-to-fuse-asymmetric-feature-maps-in | 75.3 | 81.0 | 71.2 |
| odtrack-online-dense-temporal-token-learning | 85.1 | - | - |
| 190407220 | 74.0 | 80.1 | - |
| universal-instance-perception-as-object | 85.4 | 89.0 | 86.4 |
| joint-feature-learning-and-relation-modeling | 83.9 | 88.5 | 83.2 |
| mixformer-end-to-end-tracking-with-iterative-1 | 83.9 | 88.9 | 83.1 |
| モデル 12 | 83.4 | 88.1 | 81.6 |
| target-transformed-regression-for-accurate | 78.5 | 83.8 | 75 |
| seqtrack-sequence-to-sequence-learning-for | 85.5 | 89.8 | 85.8 |
| atom-accurate-tracking-by-overlap | 70.34 | 77.11 | 64.84 |
| aiatrack-attention-in-attention-for | 82.7 | 87.8 | 80.4 |
| tracking-meets-lora-faster-training-larger | 86.0 | 90.2 | 86.1 |
| towards-sequence-level-training-for-visual | 82.8 | 87.5 | 81.4 |
| exploring-enhanced-contextual-information-for-1 | 86.3 | 90.9 | 86.1 |
| target-aware-tracking-with-long-term-context | 85.0 | 89.3 | 84.5 |
| odtrack-online-dense-temporal-token-learning | 86.1 | - | - |
| samurai-adapting-segment-anything-model-for-1 | 85.3 | - | - |
| integrating-boxes-and-masks-a-multi-object | 83.4 | 88.9 | 84.6 |
| how-to-train-your-energy-based-model-for | - | 83.7 | 73.7 |
| learning-spatio-temporal-transformer-for | 82.0 | 86.9 | 79.1 |
| exploring-enhanced-contextual-information-for-1 | 87.9 | 92.1 | 89.2 |
| mixformer-end-to-end-tracking-with-iterative-2 | 86.1 | 90.3 | 86.0 |
| swintrack-a-simple-and-strong-baseline-for | 84 | 88.2 | 83.2 |
| towards-grand-unification-of-object-tracking | 83 | 86.4 | 82.2 |
| tracking-meets-lora-faster-training-larger | 85.6 | 89.7 | 85.4 |
| siamrpn-evolution-of-siamese-visual-tracking | 70 | 79.98 | 69.38 |
| staple-complementary-learners-for-real-time | 53.59 | 60.84 | 46.72 |
| dropmae-masked-autoencoders-with-spatial | - | 88.9 | - |
| eco-efficient-convolution-operators-for | 56.13 | 62.14 | 48.86 |
| learning-historical-status-prompt-for | 84.5 | 89.1 | 83.8 |
| artrackv2-prompting-autoregressive-tracker | 86.1 | 90.4 | 86.2 |
| siam-r-cnn-visual-tracking-by-re-detection | 81.2 | 85.4 | 80.0 |