Scene labeling on CamVid 468/233 (test)
91.6Global AccuracyVGG-conv5-DAG-RNN(8)
Evaluation results
| Method | Links | ||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| VGG-conv5-DAG-RNN(8)Backbone=VGG-conv5, DAG-RNN=true, Iterations=82015.09 | 91.6 | 78.1 | — | — | — | — | — | — | — | — | — | — | — | — | |
| VGG-conv5-ENNBackbone=VGG-conv5, ENN=true2015.09 | 91 | 76.5 | — | — | — | — | — | — | — | — | — | — | — | — | |
| CNN-65-DAG-RNN(8)Backbone=CNN-65, DAG-RNN=true, Iterations=82015.09 | 88.7 | 68.3 | — | — | — | — | — | — | — | — | — | — | — | — | |
| ReSeg2015.11 | 88.7 | 68.1 | 86.8 | 84.7 | 93 | 87.3 | 48.6 | 98 | 63.3 | 20.9 | 35.6 | 87.3 | 43.5 | 58.8 | |
| SegNet2015.11 | 88.6 | 65.9 | 88 | 87.3 | 92.3 | 80 | 29.5 | 97.6 | 57.2 | 49.4 | 27.8 | 84.8 | 30.7 | 50.2 | |
| CNN-65-DAG-RNN(4)Backbone=CNN-65, DAG-RNN=true, Iterations=42015.09 | 88.2 | 66.3 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Bayesian SegNetmodel averaging=true2015.11 | 86.9 | 76.3 | 80.4 | 85.5 | 90.1 | 86.4 | 67.9 | 93.8 | 73.8 | 64.5 | 50.8 | 91.7 | 54.6 | 63.1 | |
| CNN-65Backbone=CNN-652015.09 | 84.3 | 53.2 | — | — | — | — | — | — | — | — | — | — | — | — | |
| SegNet-Basictraining=layer-wise training2015.11 | 84.3 | 62.9 | 75 | 84.6 | 91.2 | 82.7 | 36.9 | 93.3 | 55 | 37.5 | 44.8 | 74.1 | 16 | — | |
| CNN-65-ENNBackbone=CNN-65, ENN=true2015.09 | 84.1 | 58.1 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Tighe et al.year=20132015.09 | 83.9 | 62.5 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Sturgess et al.2015.09 | 83.8 | 59.2 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Ladicky et al.2015.09 | 83.8 | 62.5 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Boosting+Higher order2015.11 | 83.8 | 59.2 | 84.5 | 72.6 | 97.5 | 72.7 | 34.1 | 95.3 | 34.2 | 45.7 | 8.1 | 77.6 | 28.5 | — | |
| Boosting+Detectors+CRF2015.11 | 83.8 | 62.5 | 81.5 | 76.6 | 96.2 | 78.7 | 40.2 | 93.9 | 43 | 47.6 | 14.3 | 81.5 | 33.9 | — | |
| ReSegClass Balance=true2015.11 | 83.5 | 73.2 | 70.6 | 84.6 | 89.6 | 81.1 | 61 | 95.1 | 80.4 | 35.6 | 60.6 | 86.3 | 60 | 53.7 | |
| Supa Parsing2015.11 | 83.3 | 51.2 | 87 | 67.1 | 96.9 | 62.7 | 30.1 | 95.9 | 14.7 | 17.9 | 1.7 | 70 | 19.4 | — | |
| SegNet-Basic2015.11 | 82.8 | 62.3 | 80.6 | 72 | 93 | 78.5 | 21 | 94 | 62.5 | 31.4 | 36.6 | 74 | 42.5 | 46.3 | |
| Zhang et al.2015.09 | 82.1 | 55.4 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Bulo et al.2015.09 | 82.1 | 56.1 | — | — | — | — | — | — | — | — | — | — | — | — | |
| Bayesian SegNet-Basicmodel averaging=true2015.11 | 81.6 | 70.5 | 75.1 | 68.8 | 91.4 | 77.7 | 52 | 92.5 | 71.5 | 44.9 | 52.9 | 79.1 | 69.6 | 55.8 | |
| Tighe et al.2015.09 | 78.6 | 43.8 | — | — | — | — | — | — | — | — | — | — | — | — |