Potential solution to shaky video footage?

It seems logical and works really. If it becomes mature, maybe it will solve shaky footage underwater?

Here is briefly explain what it is and how it works:

http://www.redsharknews.com/production/item/2128-steadxp-advanced-metadata-based-video-stablisation-for-any-camera

It's working on the GH4 as well:

http://www.newsshooter.com/2015/04/11/is-it-a-gimbal-is-it-a-steadicam-no-its-steadxp-amazing-sensor-and-software-based-stabilisation-shown-working-with-panasonic-gh4/

Thought?

Looks impressive! Really clever!

I can understand how it would work for all the motion for up/down and left and right axis (like the original olympus sensor shifting) but without the footage being warped in some way I can't see how this fixes YAW and PITCH issues without reducing quality. (I've tried FCPX stabilising and it looks 'funny' if over done)

However, I too think that jittering/movement in underwater videos is one of the biggest issues and a fix like this would be cool!

Nice find!

Hi Richard, the SteadXP is using gyros, accelerometers, and GPS. If it's the gyros just like all the cell phone, then it should be very promising.

No one have seen their proprietary software yet, so besides the usual zoom/crop/rotate/scale the image frame by frame, may be they re-sample/fill in pixel?

I'm in the process of trying out a few image stabilization solutions. I don't use FCPX since I'm on PC. I tried the Warb Stabilizer in Premiere Pro, it's more consistent so far but slow and is hit and miss. Some people say the Warb Stabilizer in After Effect is better, which I don't have time to learn After Effect right now.

I learn that many image stabilization software doesn't do a good job, because they don't correct the problem of CMOS sensor so the jello effect is bad. So far only Deshaker and Mercelli V4 consider CMOS sensor.

From the videos on Vimeo, it heavily crops to achieve the results - but they are impressive. Fair play to them, showing the side by sides.

I've sent a message to them, asking why the need for Mic input and Video output (I presume for time syncing, can't understand why the mic input thou). For underwater it won't fit inside a housing, so will need a separate housing and a cabling duct.

The Unit is the same for both GoPro and any other camera, but the attachment for other cameras, has the 3.5mm inputs/outputs, which the unit clips into.

The unit fits in a standard go pro housing like a BatteryPac, so I was thinking, use that (GoPro + unit) underwater (attached to your main camera rig) and using something to sync the go pro and the other camera (clapper, hand wave or something..) you could then use the data from the unit taken with the GoPro, but for the video from your main video?

Looking forward to the Crowdfunding, I think it will be the first thing I've pledged too!

Nice job Richard, It make sense to put the SteadXP together with a goPro that slave to the main camera. If they can do it without the 3.5mm jack, that would be great since I'm using Panasonic LX100. But besides for underwater use, I think it will help tremendously for on land. On my recent trip to Anilao, I tried to do Hyperlapse from the boat, I was hoping to have a shot of the boat leaving the resort and sail to our dive site..... you can imagine the jump and bump of the banca boat.

As for their Vimeo demo, I think the fact that it's an extreme sport footage so it does require more crop in since the sport footage have more bump and jump. For UW, as long as we have decent buoyancy and not strong current (or not have 5 cups of coffee or Redbull), I would say the need of cropping in probably should be around 10 to 20% at most.

I too will jump on their crowd funding once it's open!

I can see how something like this might help with surface footage shot on a small dive boat that's bouncing on waves.

But I can't think of any problem underwater that it's going to solve. WA and MS footage doesn't have a stability problem because of the water dampening effect. I doubt if its gyros will be sensitive enough to help stabilise macro footage.

Maybe it'd be useful if shooting The Passage at Raja Ampat or Shotgun at Komodo in a full-moon current...

If it really works well, some manufacturer will buy out the IP or replicate it and build it into a future model.

I assume that, if successful, there will be plug-ins for the popular NLEs.

Regards

Peter

 

I've sent a message to them, asking why the need for Mic input and Video output (I presume for time syncing, can't understand why the mic input thou).

 

Maybe they are writing some inaudible (to humans) sync sounds onto the audio track that their NLE plug-in can read.

But I can't think of any problem underwater that it's going to solve. WA and MS footage doesn't have a stability problem because of the water dampening effect. I doubt if its gyros will be sensitive enough to help stabilise macro footage.

Even though I have my GH4 as neutral as possible (and thanks to your tips on this) it still suffers from some jittering underwater if I pan etc.

I think this is due to it not having as much mass / inertia as a larger Camcorder type rigs (Gates, etc). Maybe I just need to spend more time using it / getting used to it.

Agree that macro footage, can't see this helping much.

 

Even though I have my GH4 as neutral as possible (and thanks to your tips on this) it still suffers from some jittering underwater if I pan etc.

 

Is the housing well-balanced? Jittering can come from muscles countering balance issues.

The other tip for panning is to point your body towards the end point of the pan and then swing your arms / camera towards the start point. Start the shot still for a few seconds then pan so that your arms are moving into their natural alignment with your body rather than away from it. Your muscles relax during the pan rather than tighten up.

Regards

Peter

If I let go of the camera, it pretty much 'floats' in front of me. its about 20g negative.

I've got the Ninja2 behind the camera (via a XIT404 tripod mount) and the lights out to each side (via float arms) so the centre of gravity of the camera is better, than having the Ninja2 mounted atop.

Here is an example of my bad handy work (top is unedited)

https://www.youtube.com/watch?v=xO1B-8Afzu0

Here is an example of my bad handy work (top is unedited)

 

 

Pitch is the most difficult dimension to stablise with a "still" camera shooting video underwater. Camcorders and their longer housings are much better for pitch stability. There's no easy answer.

Maybe add another handle at the rear of your setup - right hand on housing handle, left hand on the rear handle.

Maybe add a more weight at the bottom and add buoyancy to the top - same weight, more mass.

Wrist strengthening exercises?

Practice!

Regards

Peter

Practice is probably the main one to go for [emoji6]

More practice would take care many of the shakiness. However, when you have to kick your fin to cover 20ft or so of "flyover" footage, how is it done?

There are some news and update about the SteadXP, but it's a bit disappointing since it's going to be 6-8 months out:

https://www.cinema5d.com/the-perfect-camera-stabilizer-isnt-a-gimbal-dan-chung-clinton-harn-emmanuel-pampuri-on-the-couch-ep-23-part-3-of-4/

(Warning: the interview is way boringly conducted).

So its Kickstarter time today at 18:00 (Paris time)!

http://steadxp.com/WP/wp-content/plugins/newsletter/do/view.php?id=3

Will be interesting to see what they have been up to, sounds as if there will be different variants, more than the GoPro unit, then an adaptor as shown previously...