How to time everyone's attempt in a group challenge video
Upload each person's clip and define what starts and stops the clock. Vizard Agent steps through the footage a tenth of a second at a time to find the exact frame of each start and each finish, so the times on screen are measured rather than estimated.
What do you type into Vizard Agent?
Define the two moments as precisely as you can bear to. Vizard Agent finds them frame by frame in each clip, so the difference between a comparison everyone accepts and an argument at the pub afterwards lies in how tightly you described the start and the finish. First touch and first movement are genuinely different moments.
Prompt
What does Vizard Agent actually do?
The comparison only works if everyone was timed the same way, and eyeballing it puts half a second on somebody's total. So almost the entire session goes on locating two frames per person. The timestamps are drawn onto the contact sheets, so reading a time off a frame is a matter of looking rather than counting. In the session behind this article it:
- Probed each of the five clips, extracted thumbnails, and reviewed them as a grid, rebuilt at the correct proportions, to identify the people and the movement.
- Extracted frame-by-frame images at 0.1-second intervals across all five clips.
- Generated grids at 0.3 and 0.5-second intervals with the timestamps drawn on, analysing clip one at half-second intervals and clips two to five at three-tenths of a second.
- Extracted frames around each attempt's start and end at the finest interval, and built labelled contact sheets for every successful attempt.
- Reviewed each start frame for the exact moment of first contact, and each end frame for the exact finishing position, before anything was cut.
When does this not work well?
Frame-accurate timing needs the deciding moment to be visible, and phone video shot by friends is rarely filmed with that in mind. If the start is off camera, nothing recovers it. Thirty frames a second limits precision, motion blur makes the contact frame ambiguous, and different camera angles can disagree about the moment.
Common questions
How precise can it get? To the frame. Vizard Agent steps through at a tenth of a second and narrows from there.
Can it use different definitions per person? It can, but it shouldn't. Vizard Agent applies one definition so the comparison holds.
Can I check its judgement? Yes, and you should. Vizard Agent keeps the labelled contact sheets it read the times off.