Gyros

[zak forrest]

Is there any estimate of when the first implementation of the gyro stabilization feature in software will be? either thru redcine or some other app, anyone know? it's all sinking in now while i'm shooting with the new build..

 

[Rob Ruffo]

Don't expect too much from it. It won't do anything more than post-based tools now, which easily find camera motion and can remove it. It won't fix parallax.

 

[zak forrest]

but it would save time because there is no processing/analyzing period before you can start stabilizing clips right? thats why i'm so excited about it, that and because i'm hoping it doesnt get confused by strobe lighting or tracking behind a pole like the ones i use do

can anyoen confirm/deny what i'm sayin here, or am i all excited for nothing. since the metadata is there, it can stabilize right after being loaded into the program right?

 

[Max M.]

I would like to see a horizonlevel like the 7d has, could be nice for handheld shooting..

 

[Rob Ruffo]

Zak my concern would be the resolution of recorded angles. They might be very high, or not. We simply don't know. Analysis in a tool like Warp Stabilizer doesn't really take more than 30 seconds or so.

 

[Nick Burridge]

Analysis in a tool like Warp Stabilizer doesn't really take more than 30 seconds or so.

Takes a fair lot longer than that with the 5K R3Ds I've been stabilising!

But yeah, you're right - I'm not sure the gyro data is going to be as practical for that as you might think, but a level tool/determining orientation/other wonderful applications we haven't thought of yet are all still fair game.

 

[zak forrest]

five dollars is all my mama allows me to spend

 

[charles lim yi yong]

getting the data from the gyro is much much cleaner than , video analysis . especially for quick pans where the camera points to a completely new subject or landscape.
i am really looking forward to this

 

[zak forrest]

so when is there going to be the first implementation, however crude/not-crude, im looking forward to this..

 

[Björn Benckert]

The gyro will not do a good job, or the gyros used are better than any other gyro on the market at this date...It's not exact and it does, as all gyros, have poor precision for constant movements etc. But if the gyro data stream can be feeded into a future version of syntheyes or any such pixel 3D tracking software as a helping hand to let the pixel trackers track trough jerky camera moves etc. then that will provide huge improvement to normal 3D tracking.

But my guess, based from using the motion sensors provided in the Moven movecap suite we used to have, the gyro itself is quite limited and has a to large room for jitter and errors if used as the only source of motion data.