The 120K UnrealPerson dataset mentioned in Table 1

[AlphaPlusTT]

Can you provide the 120K UnrealPerson dataset mentioned in Table 1? Is the 120K randomly selected or manual selected? Is there a large performance gap of the model pretrained with the 120K and 1.2M?

[tyz1994]

Thanks for your question. It is randomly selected in most of our experiments. For JVTC, a fixed txt file is needed. Therefore, we also provide an example txt file containing 120K image file information. (this file is also generated randomly) As for 120K v.s. 1.2M, we haven't done enough experiments. You are welcome to share your experimental results.