Thank you so much

[ztx991028]

Hello! Can you tell me where the relevant code for 6-Dof pose estimation is in the repository? How it was achieved, thank you for your reply

[aofrancani]

The 6-DoF pose is the output of the model: https://github.com/aofrancani/TSformerVO/blob/main/predict_poses.py#L80

Keep in mind that we estimate poses given consecutive pairs of frames in a windowed data. So if you have a window size of 4 (4 frames), you estimate 3 poses considering the entire windowed data. If there is an overlap between successive windowed data (sliding window with overlap), I post-process the data by taking the average of the repeated poses (they are slightly different because the estimation takes into account the entire windowed data): https://github.com/aofrancani/TSformerVO/blob/main/plot_results.py#L30