Skip to main content

Table 3 Comparison of classification accuracies (%) on Cityscapes and Foggy Cityscapes datasets

From: ML-ANet: A Transfer Learning Approach Using Adaptation Network for Multi-label Image Classification in Autonomous Driving

Method CF FC
Pedestrian Vehicle 2-wheeler Avg. Pedestrian Vehicle 2-wheeler Avg.
ML-KNN [7] 84.2 95.8 60.6 80.21 89.5 97.6 73.2 86.75
TCA [16] 75.5 94.1 51.9 73.85 76.6 93.0 55.5 74.94
JDA [17] 76.7 93.9 58.3 76.29 77.1 93.4 59.7 76.74
BDA [18] 77.9 93.9 60.3 77.35 78.3 94.0 61.6 77.97
DDC [23] 87.7 97.7 76.6 87.33 87.6 98.1 77.6 87.77
DAN [31] 90.3 99.1 86.2 91.85 95.5 99.3 92.6 95.83
Ours (ML-ANet) 93.1 99.5 91.9 94.83 96.8 99.3 94.5 96.85