Fig. 1. Convolution neural network structure

Fig. 2. Basic model of neuron

Fig. 3. Image convolution process

Fig. 4. Average pooling operation

Fig. 5. Backward propagation model

Fig. 6. Early fusion structure

Fig. 7. Late fusion structure

Fig. 8. Full connection layer fusion structure

Fig. 9. Convolution layer fusion structure

Fig. 10. Change curves of center position error and training loss function with the training steps

Fig. 11. Detection results at different algorithms. (a) Detection based on RGB images; (b) joint detection of RGB-D based on late fusion; (c) joint detection of RGB-D based on convolution layer fusion