youngwoo-yoon / Co-Speech_Gesture_Generation

This is an implementation of Robots learn social skills: End-to-end learning of co-speech gesture generation for humanoid robots.
https://sites.google.com/view/youngwoo-yoon/projects/co-speech-gesture-generation
Other
72 stars 9 forks source link

Why do 15 joints of one sample has shape [30,135]? #3

Closed chuangyu-robotics closed 4 years ago

chuangyu-robotics commented 4 years ago

in the file: ../Co-Speech_Gesture_Generation/blob/master/scripts/data_loader/lmdb_data_loader.py i run the line 69: word_seq, pose_seq, audio, aux_info = sample

I got the pose_seq has the shape [30,135]. i know 30 is the frame size of each example. and 135 is for 15 selected joints. as 135/15=9, namely each joint has 9 values. what do those 9 values mean resepctively?

Thank you in advance.

youngwoo-yoon commented 4 years ago

Nine values in a 3x3 rotation matrix. Please see process_bvh function in trinity_data_to_lmdb.py. The original paper used Cartesian coordinates of upper-body joints, but I changed to use rotation matrices since rotational angles were required in the GENEA competition.

chuangyu-robotics commented 4 years ago

I got it. Thanks a lot.