jeetv / gmvd_dataset Goto Github PK

Hi, I have read your brilliant work and your GMVD dataset, however, I wonder how you generate this dataset in GTAV? Specifically, what tools/plugins did you utilize? And also how are the cameras synchronized? How did you introduces multivew cameras into one scene? since most of the current dataset methods only have one camera perspective for each scene. Much appreciate it!

Hello, I have some confusion about the results reported in Table 7. In the explanation, you said that the results are evaluated on MultiviewX dataset but later in the paragraph, it is changed to GMVD test set. And in the table caption, it is written as GMVD test set. Which is true?