Horizon Line Estimation on ECD
90.8AUC (theta, rho)Zhai et al.
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Zhai et al.2016.04 | 90.8 | — | |
| Lezama et al.2016.04 | 89.57 | — | |
| PlacesLearning Paradigm=Classification, Loss Function=Softmax, Initialization=Places2016.04 | 83.96 | 80.45 | |
| SalientLearning Paradigm=Classification, Loss Function=Softmax, Initialization=Salient2016.04 | 82.62 | 80.11 | |
| ImageNetLearning Paradigm=Classification, Loss Function=Softmax, Initialization=ImageNet2016.04 | 82.28 | 82.99 | |
| Best (Regularized Huber)Learning Paradigm=Regression (regularized w/ classification), Loss Function=Huber, Initialization=Best Classification Network2016.04 | 81.79 | 82.55 | |
| Best (Regression Huber)Learning Paradigm=Regression, Loss Function=Huber, Initialization=Best Classification Network2016.04 | 81.19 | 81.85 | |
| Best (Regularized L2)Learning Paradigm=Regression (regularized w/ classification), Loss Function=L2, Initialization=Best Classification Network2016.04 | 79.24 | 82.89 | |
| RandomLearning Paradigm=Classification, Loss Function=Softmax, Initialization=Random2016.04 | 78.63 | 77.17 | |
| PoseNetLearning Paradigm=Classification, Loss Function=Softmax, Initialization=PoseNet2016.04 | 78.36 | 77.77 | |
| Places (Regression Huber)Learning Paradigm=Regression, Loss Function=Huber, Initialization=Places2016.04 | 76.72 | 76.72 | |
| Best (Regression)Learning Paradigm=Regression, Loss Function=L2, Initialization=Best Classification Network2016.04 | 76.65 | 76.59 | |
| Places (Regression)Learning Paradigm=Regression, Loss Function=L2, Initialization=Places2016.04 | 71.43 | 69.7 |