GPS/INS timestamp and fisheye camera timestamp not synchronized

[nkwangyh]

I've found a few sequences don't have their timestamps of GPS/INS and fisheye camera synchronized. For example the sequence http://robotcar-dataset.robots.ox.ac.uk/datasets/2014-06-23-15-36-04/. It seems that in the sequence when GPS/INS has already turned around at the end of the sequence, the fisheye camera is still a few seconds from the turn. But most of the sequences are well synchronized.

[mrgransky]

could u plz explain how u find out each frame in one dataset corresponds to a frame in another dataset using gps/ins info? timestaps provided for stereo.timestamps / mono.timestamps are not the same with gps.csv or ins.csv !