Hi,
Thanks for your great work and congratulations on CVPR '26.
I'd like to ask about how do to preprocess the augmented PartNet-Mobility subset from SINGAPO, and whether the code will be released.
More specifically, SINGAPO provides JSON data of each object and its joint parameters, but they are unnormalized and it's hard without the augmented URDFs too. Hence, to train models, how do you normalize the coordinates and prepare stuff from that set?
Thank you!
Hi,
Thanks for your great work and congratulations on CVPR '26.
I'd like to ask about how do to preprocess the augmented PartNet-Mobility subset from SINGAPO, and whether the code will be released.
More specifically, SINGAPO provides JSON data of each object and its joint parameters, but they are unnormalized and it's hard without the augmented URDFs too. Hence, to train models, how do you normalize the coordinates and prepare stuff from that set?
Thank you!