Dear all,
Here're some frequent questions the participants will ask. I think it might be helpful to list them for all the participants.
1. Q: Whether I can use synthetic data for this challenge?
A: Yes, any additional dataset can be used for this challenge. However, there're some points that need to be considered: a. The .stl file only includes XYZ information rather than RGB information, which may cause some problems when 3d rendering. Some realistic colors and materials need to be added for synthetic data. b. the pose of synthetic data must correspond to the original dataset. To check the pose axis of the 3D model, you can use show_axis() function in visualization.py
2. Q: Whether all the data is collected in sequences?
A: Yes. Both training and testing data are organized in sequences which means it contains temporal information. However, the data is extracted from multiple separate video clips rather than a whole long video. A video clip might include 100 frames in average.
More questions will be listed for discussion.
Please feel free to ask any questions during the challenge : )