Multimodal Multiobject Tracking by Fusing Deep Appearance Features and Motion Information

Table 1

The architecture of the used CNN.

Name	Patch size/stride	Output size
Conv 1	3 × 3/1	32 × 128 × 64
Conv 2	3 × 3/1	32 × 128 × 64
Max pool 3	3 × 3/2	32 × 64 × 32
Residual 4	3 × 3/1	32 × 64 × 32
Residual 5	3 × 3/1	32 × 64 × 32
Residual 6	3 × 3/2	64 × 32 × 16
Residual 7	3 × 3/1	64 × 32 × 16
Residual 8	3 × 3/2	128 × 16 × 8
Residual 9	3 × 3/1	128 × 16 × 8
Dense 10	—	128
Batch and normalization	—	128