HumanML3D
HumanML3D copied to clipboard
EricGuo5513
Question about Dataset
Hi, thank you very much for the great work. I notice that only a subset (18/24) of AMASS is used in HumanML3D. I am wondering why selecting these 18 subsets but not all of them, for example, GRAB.
Would you please give some insight on this? Thanks a lot!
Hi, at that time, only these 18 subsets are available. The remaining 6 datasets are added into AMASS after our work.
On Sun, 15 Oct 2023 at 13:31, Coral @.***> wrote:
Hi, thank you very much for the great work. I notice that only a subset (18/24) of AMASS is used in HumanML3D. I am wondering why selecting these 18 subsets but not all of them, for example, GRAB.
Would you please give some insight on this? Thanks a lot!
— Reply to this email directly, view it on GitHub https://github.com/EricGuo5513/HumanML3D/issues/96, or unsubscribe https://github.com/notifications/unsubscribe-auth/AKRYNBYVQ347ESNPGFYWIITX7Q2YZANCNFSM6AAAAAA6BGYRSI . You are receiving this because you are subscribed to this thread.Message ID: @.***>
Thank you very much for the fast reply. I am wondering if it would improve the performance of the language-conditioned model if the remaining 6 subsets are processed and added to HumanML3D. I am wondering if these 6 subsets will bring much more of variety to the current dataset
Hi, I guess the dataset is the larger the better. Probably we need to look into these other 6 subsets, to see they would introduce more diversity or they are something similar in existing humanml3d.