Human3.6M D3 angles

[sadegh-aa]

Hi,

Somewhere in your paper, you mentioned that H3.6M has 54 independent joint angles in 3D space. Are these correspond to 18 joints? If possible, I really appreciate if you could please give me a link to the data that contains these 54 joint angles (with their corresponding joint index) per sequence.

I was looking at H3.6M both on its website and the link you've mentioned in your repo. The values are different in different sources. Moreover, in the dataset you've used, there are 99 values per frame, and you select 54 of them. However, I'm not sure what are these 54 values per frame. Are they correspond to 18 joints in 3D angle space? if yes, would you please let me know the triplet (x, y, and z) indices correspond to each joint?

Thanks

[una-dinosauria]

Hi Sadegh,

Somewhere in your paper, you mentioned that H3.6M has 54 independent joint angles in 3D space. Are these correspond to 18 joints?

Not necessarily. Some joints have only one or two degrees of freedom, so it's not necessarily an n/3 relationship.

If possible, I really appreciate if you could please give me a link to the data that contains these 54 joint angles (with their corresponding joint index) per sequence.

I believe the code that we provide filters out dimensions that do not move (ie, joints that do not bend), and the resulting 54 numbers are the degrees of freedom we are left with.

I was looking at H3.6M both on its website and the link you've mentioned in your repo. The values are different in different sources. Moreover, in the dataset you've used, there are 99 values per frame, and you select 54 of them. However, I'm not sure what are these 54 values per frame. Are they correspond to 18 joints in 3D angle space?

The first three values are the global translation of the person. The rest correspond to joint angles of the kinematic tree of a person.

If yes, would you please let me know the triplet (x, y, and z) indices correspond to each joint?

We have code to convert form kinematic tree to (x, y, z) pairs so that we can plot the sequences in: https://github.com/una-dinosauria/human-motion-prediction/blob/master/src/forward_kinematics.py#L14

I suggest thoroughly going through our forward kinematics code, our visualization code and familiarizing yourself with https://en.wikipedia.org/wiki/Forward_kinematics.

Finally, if you have further questions, please direct them to the authors of the dataset (http://vision.imar.ro/human3.6m/description.php).