Shooting video from a motorbike problems

[Doctor Keo]

Hi,
I shot some video from a Panasonic V720 camera, mounted on my motorbike, on the fully automatic setting, but the video had far too much shaking in it. I tried the various setting on PowerDirector 13 stabiliser (the new version has really helped some of my old video) but that video seems beyond repair. I think if I use the manual shutter setting to up the shutter speed it would improve future video. Does anyone have a suggestion of what speed to set the camera for bright sunlight - I don't use the motorbike unless the weather is good I can set the speed right up to 1/8000.

Thanks
Doc

[smrpix]

The shutter speed has nothing to do with stabilization. Interacting with the iris and gain It affects exposure, depth of field and motion blur -- the manipulation of these provide subjectively more pleasing images -- it's a creative choice, so experiment.

The only things that affect stabilization is the optical and electronic stabilizer, and to some extent the wideness of your lens. If it's insufficiently smooth you need to add stabilizing hardware like a steadicam.

Post-production stabilization can be very effective if your original footage isn't too wild. Faster shutter speeds can actually work against you in this situation.

[Doctor Keo]

Hi - Thanks for your comments - the reason I still think that increasing the shutter speed will help is that some of the objects in the frames seem to be stretched, which makes me think that while the lens is open for each frame the camera has moved. Being able to 'freeze' the scene with a faster shutter speed and using a stabilizer to resync the frames within video I'm hoping to resolve the problem.

Another thing that isn't made clear in the Panasonic V720 instruction is whether once i have set the shutter speed on the camera the camera will automatically set the aperture

Thanks Doc

[Cornucopia]

HOW do you have your cam mounted?

Scott

[Doctor Keo]

The camera mount is just a cheap grip to the handlebars.

Doc

[smrpix]

Another thing that isn't made clear in the Panasonic V720 instruction is whether once i have set the shutter speed on the camera the camera will automatically set the aperture

Normally the automatic controls adjust and compensate for whatever you've set manually.

Being able to 'freeze' the scene with a faster shutter speed and using a stabilizer to resync the frames within video I'm hoping to resolve the problem.

..Until it confuses the sharp edge of building C for the sharp edge of building A or gets confused by aliasing lane stripes...

[Cornucopia]

The camera mount is just a cheap grip to the handlebars.

Doc

That's a BINGO!

You probably need something that is more specialized. It needs to have mass (to lower the freq of vibration), needs to be nearest the "fulcrum" of your vehicle (to minimize the torque & amplitude of vibration) - such as the frontspiece of the frame that your front fork goes through (the "head tube"?), and needs to have it's own damping to isolate it further from the native vibrations/shocks of the bike. And of course, TIGHT.

Rubberized/cushioned bracing, with possible gimballed counterweighting (not to mention gyros) would further improve the isolation from vibration.
But you must realize - you are on a motorbike! Even the best camera on the best mount is going to have problems. You have the vibration of the motor, the vibrations of your steering and the shocks/vibrations of the contact with the road all while perched on something with only 2 wheels (points of stability).

Scott

[Doctor Keo]

Thanks i'll have a look at improving the mount.
Doc

[newpball]

Hi,
I shot some video from a Panasonic V720 camera, mounted on my motorbike, on the fully automatic setting, but the video had far too much shaking in it. I tried the various setting on PowerDirector 13 stabiliser (the new version has really helped some of my old video) but that video seems beyond repair. I think if I use the manual shutter setting to up the shutter speed it would improve future video. Does anyone have a suggestion of what speed to set the camera for bright sunlight - I don't use the motorbike unless the weather is good I can set the speed right up to 1/8000.

Thanks
Doc

You can't really solve that problem.

It is probably best to mount a GoPro on your helmet and run image stabilization in post.

[redwudz]

I'd agree about the helmet mount. Make use of your neck and body as the primary shock absorber.
It can work quite well. And add some foam rubber around the camera to remove a bit more shake.

From there it can get a bit complicated for a really smooth mechanical mount on a motorcycle.
I would rely on software stabilization as a very last resort.

EDIT: Take a look at a few helmet cam videos on youtube. Most are reasonably smooth.
Most are likely using a GoPro camera cause it's easy to mount and relatively cheap.

[Cornucopia]

But again part of GoPro's ability to handle this is the very wide angle that GoPros capture. Their normal minimal FOV is >120degrees. At that wide of an angle, the blur due to vibration traverses a much smaller number of pixels compared to your normal camera's FOV (~30-80degrees).

Scott

[budwzr]

Correct-o Mundo on that wide field of view. Maybe a fisheye lens on your cam would work.

[darkknight145]

" the reason I still think that increasing the shutter speed will help is that some of the objects in the frames seem to be stretched."

This is because the camera you are using is a CMOS camera, CMOS chips are scanned, so if any movement happens during the scan you get image distortion.
CCD chips are not scanned so it's not a problem.
Of course as others have said, get a better mount to try and help remove camera vibration, even mounting the camera on your helmet would reduce vibration as your body will adsorb it. But there is a safety issue using a helmet cam.

[newpball]

But there is a safety issue using a helmet cam.

I must miss something, how can a 3.5 oz camera be a safety problem?

[darkknight145]

But there is a safety issue using a helmet cam.

I must miss something, how can a 3.5 oz camera be a safety problem?

Schumacher's injuries were possibly caused by his GoPro helmet cam according to some articles!
Plus you have to get proper fittings and attached properly.

[newpball]

Schumacher's injuries were possibly caused by his GoPro helmet cam according to some articles!

The reporter who claimed this quickly recanted his "findings".

http://www.reuters.com/article/2014/10/15/us-shumacher-gopro-idUSKCN0I42S320141015
http://www.forbes.com/sites/ryanmac/2014/10/15/gopro-may-pursue-legal-action-against-j...h-accusations/

[smrpix]

I must miss something, how can a 3.5 oz camera be a safety problem?

Theoretically -- falling on the the camera can apply more force to a single point on the helmet. The helmet is less able to redistribute the impact, thereby causing greater injury. The camera acts as a blunt spike in this scenario.

Whether that's that's actually happened with Schumacher, we'll probably never know.

[newpball]

Theoretically -- falling on the the camera can apply more force to a single point on the helmet. The helmet is less able to redistribute the impact, thereby causing greater injury. The camera acts as a blunt spike in this scenario.

Oh boy.........sigh!

I strongly suggest that if someone is worried about this he should not ride a motorcycle at all as a motorcycle has about 6 times the fatality rate over a car. Or just stay at home, with a helmet on of course!

[Doctor Keo]

Thanks for your suggestions and advice. After trying various shutter settings and different stabilizers the forum is correct I was unable to solve the unwatchable video - I was able to improve the video - but even then it is by 'cheating' somewhat (driving slow and steady) and now I have broken the bike mount which brings my little experiment to end end for the moment. It looks like I will have to put a Go Pro (and mount) on my next Christmas list.
Thanks again

[pippas]

This is because the camera you are using is a CMOS camera, CMOS chips are scanned, so if any movement happens during the scan you get image distortion.
CCD chips are not scanned so it's not a problem.

darknight145 has hit on the root of the problem - and even the Go Pro uses a CMOS sensor, so using that won't totally solve the problem.

This video shows an extreme example, but it makes the point... https://www.youtube.com/watch?v=QwajiIzGQVg

[Cornucopia]

It's a shame that example doesn't show any comparison of equals: equal camera tier, equal optics, equal road conditions, or even equal FOV. Yes, some/many CMOS sensors have the "jello" effect, but this was exaggerated - probably to belabor a point.

I'm sure certain people will come along and deride CMOS and advocate the OP chucking his current gear and getting 3CCD, but the number of good quality 3CCD cams that are reasonably priced can be counted on 2 hands at most.

Fact is, there are plenty of cams with CMOS that should work quite well if mounted/isolated/stabilized correctly. Just as there is no guarantee that a 3CCD cam would be impervious to shake/vibration (it would only be impervious to the jello effect). And good mounting often costs (much?) less than completely re-investing in new cam gear.

Scott

[Doctor Keo]

[QUOTE=pippas;2373076]

This video shows an extreme example, but it makes the point... https://www.youtube.com/watch?v=QwajiIzGQVg

This is a good example (although I have far more shake, due to the roads in Cyprus being terrible) of the video results i get. I have found an old video camera which is CCD (Panasonic NV-EX1) but there so many problem with that 1. Battery. 2. Mini Tapes 3. Transfer video AND it is weighs twice as much as my V720 - that is even if i can repair the camera mount!

[pippas]

It's a shame that example doesn't show any comparison of equals: equal camera tier, equal optics, equal road conditions, or even equal FOV. Yes, some/many CMOS sensors have the "jello" effect, but this was exaggerated - probably to belabor a point.

I'm sure certain people will come along and deride CMOS and advocate the OP chucking his current gear and getting 3CCD, but the number of good quality 3CCD cams that are reasonably priced can be counted on 2 hands at most.

Fact is, there are plenty of cams with CMOS that should work quite well if mounted/isolated/stabilized correctly. Just as there is no guarantee that a 3CCD cam would be impervious to shake/vibration (it would only be impervious to the jello effect). And good mounting often costs (much?) less than completely re-investing in new cam gear.

Scott

That video was originally posted in 2009, HERE but the comments on that are now disabled .
IIRC, the OP got fed up with getting so many comments about the testing set up. I think he used 2 Sony camera from that time -one CCD, one CMOS - and the mounting arrangements were identical. But a lot of folk didn't believe that, and kept posting that they didn't believe him! So he disabled comments.
As I say, it's an extreme example, and I think these earlier CMOS sensor 'scan' speeds were probably slower, making the problem worse.

Another sample from that earlier time HERE also shows how different these earlier CMOS camcorders were from their CCD equivalents at that time. And how unsuitable they were in high vibration situations!

But you can't escape the fact that rolling shutters will tend to give some 'jello' distortion in high vibration situations, and should be avoided if possible. After all, 'jello' type distortion is not really something you can deal with effectively in post..

As time goes by we shall hopefully see global shutter CMOS sensors appearing in cheaper cameras (AFAIK, the Blackmagic Production Camera is currently the cheapest global shutter CMOS camera available.... so there's a way to go yet before the feature reaches us normal mortals!)

But for most situations, the benefits of CMOS sensors do seem to outweigh the problems -- and most folk never notice rolling shutter anyway!

[Charliebrm]

Still too early in the season for riding here but this year I am going to experiment with a camera mount attached to a tank bag - using the bag for its height and its damping effect. Maybe fill it with towels or other soft contents. My motorcycle is a 1100 cc, 4 cylinder Yamaha.

Last year I briefly tried a Canon Vixia HFR50 on a rigid handlebar mount and it was useless - all jello effect. I did switch settings in the menu for speed/shutter, etc. No help.

In my car on the same roads at the same speed I can get good video just from holding the camera on the dashboard with one hand, so it's not the camera at fault.

I would be limited to shooting through the windscreen with this setup so I am researching the powered gimbal mounts used primarily for roto-copters. Some of these gimbals will take any light camera with a standard threaded tripod mount.