Dear authors, thanks for your amazing work. I am trying to follow your method, and have some questions about \data usage:
1) Which subjects in CAPE did you use for the reported metrics? Are they beyond the publically available 20 raw scan sequences of 4 subjects in CAPE?
2) You mentioned in the paper that "the motion sequences are randomly split into training (80%) and test (20%) sets". Do you mean "80% frames" or "80% sequences"? Is it true that at least a part of each sequence is used for training?
We use all subjects you can find in CAPE dataset. Yes they are beyond the generally available raw scans. Apply through CAPE webpage and see if you can get more raw scans.
The first 80% sequences. The rest 20% seqs are completely left out for testing.
The training code of SCANimate is available, you can train on your own data to do comparisons based on your need. Probably you need to change line 108 in the configs/example.yaml file to better suit your data. In general 1000~2000 would be good for sequences in a total of ~100 frames.
Dear authors, thanks for your amazing work. I am trying to follow your method, and have some questions about \data usage:
1) Which subjects in CAPE did you use for the reported metrics? Are they beyond the publically available 20 raw scan sequences of 4 subjects in CAPE?
2) You mentioned in the paper that "the motion sequences are randomly split into training (80%) and test (20%) sets". Do you mean "80% frames" or "80% sequences"? Is it true that at least a part of each sequence is used for training?
Thank you!