Visual Emotion Distribution Learning on Twitter_LDL (test)
0.22Chebyshev DistanceEmotion Circle (Ours)
Evaluation Results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Emotion Circle (Ours)Category=CNN-based, Backbone=ResNet-50, Loss Function=Progressive Circular loss2021.06 | 0.22 | 0.84 | 0.76 | 0.44 | 0.89 | 72 | 1.5 | 78 | |
| E-GCNCategory=CNN-based, Backbone=ResNet-502021.06 | 0.24 | 0.85 | 0.78 | 0.46 | 0.87 | 70 | 3.8 | 76 | |
| JCDLCategory=CNN-based, Backbone=ResNet-502021.06 | 0.25 | 0.83 | 0.76 | 0.53 | 0.85 | 68 | 3.2 | 76 | |
| SSDLCategory=CNN-based, Backbone=ResNet-502021.06 | 0.25 | 0.84 | 0.76 | 0.51 | 0.86 | 69 | 2.8 | 77 | |
| DLDLCategory=CNN-based, Backbone=ResNet-502021.06 | 0.26 | 0.84 | 0.77 | 0.54 | 0.83 | 65 | 5.2 | 73 | |
| ACPNNCategory=CNN-based, Backbone=ResNet-502021.06 | 0.27 | 0.84 | 0.78 | 0.58 | 0.84 | 64 | 6.5 | 74 | |
| AA-kNNCategory=Algorithm Adaptation (AA)2021.06 | 0.28 | 0.58 | 0.41 | 3.89 | 0.82 | 66 | 5.8 | 73 | |
| SA-IISCategory=Specialized Algorithms (SA)2021.06 | 0.28 | 0.86 | 0.79 | 0.64 | 0.82 | 63 | 8.5 | 70 | |
| CNNRCategory=CNN-based, Backbone=ResNet-502021.06 | 0.28 | 0.85 | 0.76 | 0.67 | 0.82 | 58 | 6 | 74 | |
| SA-CPNNCategory=Specialized Algorithms (SA)2021.06 | 0.36 | 0.85 | 0.78 | 0.85 | 0.75 | 56 | 9.3 | 70 | |
| AA-BPCategory=Algorithm Adaptation (AA)2021.06 | 0.37 | 0.89 | 0.84 | 1.19 | 0.71 | 59 | 10.8 | 72 | |
| SA-BFGSCategory=Specialized Algorithms (SA)2021.06 | 0.37 | 0.89 | 0.84 | 1.19 | 0.71 | 57 | 11.2 | 57 | |
| PT-BayesCategory=Problem Transformation (PT)2021.06 | 0.53 | 0.85 | 0.77 | 1.31 | 0.53 | 40 | 10.7 | 45 | |
| PT-SVMCategory=Problem Transformation (PT)2021.06 | 0.63 | 0.91 | 0.88 | 1.65 | 0.25 | 21 | 13.8 | 40 |