Multiple Feature Fusion Based on Co-Training Approach and Time Regularization for Place Classification in Wearable Video

<table>Comparison of single (BOVW, CRFH, and SPH) and multiple feature (SVMDAS, SimpleMKL) approaches for different supervision levels. Plot of the accuracy as a function of the supervision level (IDOL2 dataset, random setup).</table>

Advances in Multimedia

fig6

Figure 6

Figure 6: Multiple Feature Fusion Based on Co-Training Approach and Time Regularization for Place Classification in Wearable Video 