Who can explain the meaning of format in './data/LSVRC2015/ImageSets/VID_train_15frames.txt' and'./data/LSVRC2015/ImageSets/VID_val_videos.txt' about flow-guided-feature-aggregation HOT 4 OPEN

Well, as far as I decipher:

The first digit is always 1 and it denotes that this frame is present (and not absent that is). Not really helpful indead.

The other 2 together reveal the sampling of all frames from each video. The intention is to keep exactly 15 frames from each video. The final number is the total number of frames of this particular video and the penultimate number is id of the sampled frame.

For example for the 1st video (ILSVRC2015_train_00000000) the 1st 3 lines are:

train/ILSVRC2015_VID_train_0000/ILSVRC2015_train_00000000 1 10 300
train/ILSVRC2015_VID_train_0000/ILSVRC2015_train_00000000 1 30 300
train/ILSVRC2015_VID_train_0000/ILSVRC2015_train_00000000 1 50 300

which means 300 frames are present and the 1st sampled frame is the 10th one, the second the 30th etc.

from flow-guided-feature-aggregation.

@PCML-WangZhe
Have you known the meaning of the last three numbers?I also have the same question!

from flow-guided-feature-aggregation.

@ Now I want to train own data which have 'person' label .But I do not know how to do my data as author's train data and to download my data using author's code.Can you tell me how to do? Thank you very much.

from flow-guided-feature-aggregation.

@eypros
Hi~ About the penultimate number , I notice they have different intervals when TotalFrameNumber changes. Do you know how the penultimate number are set(calculated)?
Thanks!

from flow-guided-feature-aggregation.