Work in progress (not cleaned up)
Multi-layered LSTM was used for the change of pose regression.
Dropout.
The results of the training are shown on the graph below:
The validation set loss was equal to 0.001 (1 mm)
Work in progress (not cleaned up)
Multi-layered LSTM was used for the change of pose regression.
Dropout.
The results of the training are shown on the graph below:
The validation set loss was equal to 0.001 (1 mm)