Multimodal Multiobject Tracking by Fusing Deep Appearance Features and Motion Information

Figure 5

Some samples in the VERI dataset. From top to bottom are cars, trucks, and buses.