Question on training scheme

[tarun005]

Hi,

In section 5.1, when you compare results for 8x interpolation, does this mean that the interpolation is being performed from (1000/8=)125 FPS to 1000 FPS? And what does 8x mean on Adobe dataset, does it mean 30-240FPS?

Also, from your dataset, even through the videos might be 1000FPS, you still seem to sample them at 25 FPS, this is because a one second clip from your dataset video is giving 32 frames, while it should give around 1000 frames for 1000 FPS videos. Can you clarify this part?

[hjSim]

Q1. does this mean that the interpolation is being performed from (1000/8=)125 FPS to 1000 FPS? A1. For X-TEST, please refer to right side column of Section 3 in the main paper (page 3) : "Each scene for X-TEST simply contains one test sample that consists of two input frames in a temporal distance of 32 frames and approximately corresponds to 30 fps. The test evaluation is set to interpolate 7 intermediate frames, which results in the consecutive frames of 240 fps."

Q2. what does 8x mean on Adobe dataset, does it mean 30-240FPS? A2. Yes.

Q3. ... Can you clarify this part? A3. We construct both X-TRAIN and X-TEST clips at 1000fps. For X-TEST , however, we only evaluate intermediate frames 4, 8, ..., 28th frames when two input frames 0, 32th are given for the simplicity (approximately 30fps -> 240fps, which is x8 multi-frame interpolation). For X-TRAIN, we fully exploit the dense 1000fps video clips as training samples.