I’ve been attempting to make the PVC Steadicam featured on this site. It looks pretty hacked up, but the results look pretty impressive, and it seems like a relatively easy build.
I’ve been out and got most of the tools and materials that i need and after about 30 minutes i had got surprisingly further than i thought i would. The gimbal is in place and the threaded bar and wooden weight is attached. The gimbal seems to work pretty well so far although i couldn’t seem to get the same thickness of PVC in the example. I think this is going to be my major problem. For now though it seems to be holding together ok.
The inner PVC coupling that hugs the skate bearing isn’t as tightly huggy as i would like and i think i’m going to have to come up with a better way to attach these two pieces together, as this join is where most of the weight is going to be concentrated. I’m thinking bigger nuts that cover both the PVC and the bearing to hold them i place.
The hardest part of this job is going to be building a mount onto the top of the threaded rod to attach the camera to. The example site doesn’t have instructions on how he built his. I’m thinking affixing some sort of tripod head to a piece of wood is going to be the go. I’ll keep this post updated with results as i go.
*************update wed 4th************* Just got a bunch of washers to use as weights for the slat of wood at the base. I’ve picked up about 38 of them, which worked out at about $15 including the bolts. This has been the most expensive part of the whole thing so far. All the PVC only cost me about $8.
I’m not sure they’re actually going to be heavy enough, but we’ll see. I’m assuming they will need to be *about* the same weight as the camera, and they’re definitely not that, but then again they’re at the *long* end of the rod, so maybe that makes they’re weight proportionally more than the shorter, top end. This is what the dude used in the reference site by the looks of it too – fender (mudguard) washers.
I’m a self-confessed and utterly obsessed early adopter, band-wagon-jumping, gear-head. This is something that im try to settle down on, but after being a Pentax guy for about 4 years I’ve finally sold / traded in all of my gear and switched to a Canon system. This is where i’m going to stay, safe in the comfort that i’m using a really good quality system and i’m compatible with all my friends. This is my story…
I originally picked Pentax when i was working at Sputnik, we had CR Kennedy as a client and my CD and i got a great deal on the K10D back in the day. It was touted as the best camera Pentax had made in 20 years and it probably was. It did everything i needed at the time and was a nice slide back into the world of SLR where i had been vacant since my days as a photography assistant for Nick Osborne at Osborne Imaging. After an unfortunate accident at a VJ Xmas party the K10D was retired for a new model, the Pentax K20D which was a bit faster, a bit bigger and a bit higher resolution. Then came the dawn of the Video DSLR and Pentax had an answer for me there with the K-7. After getting well-stuck into this scene, investing in a few lenses and being very comfortable with using the Pentax system for many years i was pretty happy… but something incide me was never completely convinced that i was walking the right path.
Everywhere around me people were using these fancy Canons and Nikons, and although i was proficient with my Pentax, i couldn’t help being just a bit jealous of their gear (and their ability to share gear with one another). One of the reasons i got the Pentax K-7 was it’s ability to shoot 720p HD video, but my biggest gripe with the K-7 was that although you could manually control the aperture, it would automatically ramp the shutter speed to keep the light consistent once you set it. This caused a few problems; Not being able to maintain one set exposure when moving in and out of light sources seemed a bit amateur and then not being able to avoid flicker from the refresh speed of lights when they matched the camera was the killer. I have some tungsten lights that i use for interview and portrait setups and the K-7 always seemed to match the refresh rates and caused a really nasty flicker unless i set the camera to automatically set the iso in movie mode (when then changed the exposure when they lights changed). For me, the tipping point was seeing all the great results my friends were getting with their Canons, It was all a bit much. I’ve now taken the plunge.
I decided to sell all of my Pentax gear, 2 camera bodies, 4 lenses (and a trusty old 3 chip MiniDV camera that i’ve had for ages) and just start buying Canon. After a months worth of trading between friends, selling on Gumtree.com.au and eBay i’m finally left with no more Pentax in my life, and a week ago i headed down the road and picked up a Canon 7D kit with the 15-80mm lens, a 30mm f1.4 Sigma, a 50mm f1.4 sigma and a Canon Speedlite 580EXII flash. I feel like i need a longer zoom coming from having an 18-200 which granted wasn’t the best quality but was a good all-rounder, but the Canon L Series 70-200mm seems a little decadent after getting all this stuff. Xmas is just around the corner though. All said, done and purchased, I feel like a new man.
First of all, the 7D feels like a big solid chunk of quality. It’s a fair bit heavier than the Pentax, but i don’t mind that as i’m totally strong. It’s taken a while to get used to a new layout and all new controls. The Pentax had a great still-shooting setting (P) which would allow you to set an aperture and it would auto-adjust the shutter, or set the shutter and it would auto-adjust the aperture. A bit like a combo of av and tv modes on the fly. The canon doesn’t seem to do this in the same way, but i’ve gotten used to shooting full manual now, as the metering is very accurate and the controls are nice to use. Below is my favourite shot from Friday night, this guy is the cleaner at work, he’s got a lovely smile and a great face. I have no idea who that crazy scot is behind him though.
I’ve never really shot with an external flash before, but have found myself needing one more and more. After playing around in the lounge room, i’ve resulted in a bouncy way of achieving nice, natural light while using the flash. This purchase really was inspired by the amazing results that my friend Ty gets with his flash, he’s opened my eyes and i’m totally on board now.
One massive advantage that the Canon’s have is that the movie mode is fully integrated into the camera and so instead of switching to 1 movie mode that does everything for you, the movie mode can be enabled on any other shooting mode than the camera offers. This is an absolute godsend and very intuitive, allowing a bucketload of versatility. There is a dedicated switch on the back of the camera to switch between stills and movie mode and a button for enabling live view in stills mode, or on/off rec in movie mode. You can still hit the shutter and take a still shot while shooting video though. Dave has been getting some amazing results out of his 5DII and we’ve been going back and forth talking about the merits of Canon’s range and the available lenses. Obviously the best advantage of shooting video on cameras like this is the fact that you can achieve beautiful images if you’ve got the right lenses and know how to use them, getting gorgeous short focus shots that look more like film than the movie mode of a stills camera. One of the main disadvantages is the shakiness that is hard to avoid when hand-holding these camera. Places like Red Rock Micro and Zacuto (among others) have some great gear attempting to alleviate this issue and i’m hoping to try out some of that stuff one day. But for now, i’ll just shoot pancake sunday(below) in slow motion and that should hopefully make my jittery hands look a bit less jittery.
Just a little video that Shaun, Ryan and i put together for 2009’s St. Kilda City Presentation Night. We were Premiers (and champions actually) this year, and generally that spurs on a bit of effort for presentation night. The last time this happened we re-did a scene from 300 which i really need to dig up. It was all just stills manually keyframed onto the actor’s faces. It was pretty hacked together. And now here’s presentation night video 2.0, with shot video, motion-tracked onto actor’s heads. It’s still totally shit, but it was funnier working with video rather than stills this year.
We put it together over the space of a couple of nights. Along with some other stuff that i can’t upload anywhere because it’ll get pulled for copyright infringement. Namely the music video for Def Leopard’s ‘Rocket’ with a dude from the club’s head flashing onto the screen every time the word rocket is sung. It also has a credit roll of all the players and administration involved with the Premiership winning team.
Had we had a bit more time it would have been fun to clean this up a lot more and do it properly. But it is what it is. And here it is…
Somehow (and i have absolutely no idea how) the first time that i exported the audio from the trailer, it didn’t export Brad Pitt’s voice with it. I have no idea why, but it was pretty handy to be able to just plonk in the new audio track that we (very badly) ADR’d and still play the backing track underneath it. I needed some of the other voices from the trailer in at other times and when i exported the audio a second time it included them. If anyone can indulge me as to why this might have happened it’d be greatly appreciative. Maybe it was a left / right channel thing? Or maybe the vocal and effects track was actually separate in the original quicktime, but it worked out very handy for me for this purpose.
Barbie is old… Bratz are slutty… Introducing Liv. Real Girls, Real Life.
“Liv In My World” [ www.livinmyworld.com.au] is a newly launched website, designed and developed at Visual Jazz to showcase a new range of dolls distributed by Funtastic in Australia called Liv. Katie, Daniella, Sophie and Alexis are the girls that make up the complete range.
The website features Katie, one of the girls in the Liv range. You can outfit her with about 60 different combos of clothes and hairstyles and see her in augmented reality by printing out and holding up a marker to your webcam.
The site has had a pretty quick turn-around to meet a launch date for the sale of the dolls in store, it’s been a technically challenging job to work on with us trying a few different techniques for various features. We’ve now got a really great pipeline worked out for working with 3D collada models in Papervision3D and being able to test them in-browser with all their textures etc… It’s now very easy for our designers and 3D artists to update the models and textures and test them in the engine without having to engage flash. This is pretty important because of the intricacies of the Papervision engine. It seems to have some pretty bad clipping issues when separate planes are too close to one another, we first encountered this when working on the 3D submarine model for the ‘Ocean Recon‘ project. It’s a great engine that works well, but there’s some pretty serious competition out there and competition is only going to make thing better and cooler for 3D on the web.
Liv In My World is the 2nd augmented reality project that we’ve launched in a week (alongside Flash Forward Experience). It’s been pretty funny sitting at my desk going back and forth between these 2 jobs for the last month, almost as if i was some sort of dedicated AR specialist / fan-boy / loser… I’m not, honest. It’s actually been really good to get into this tech and realise that a really big part of getting this stuff right is the path that the user must follow to actually initialise something like this.
For this project it was important not to alienate all of the users that don’t have cameras connected to their phones. We gave them a way to get some sort of experience without it, by being able to play around with Katie by dressing the model in the website interface, young girls could still get a sense of the look of the doll and interact with the customisability of her. Only if they wanted to bad enough and were well enough equipped would they use the augmented reality component. And the demonstration steps were very blatant and (hopefully) very easy to understand and follow. Getting this right is in my opinion absolutely necessary to the success of an AR job. It’s a pretty foreign thing to do for most people (print out a weird picture and hold it up to your webcam) and it’s easy to forget working with this stuff all the time that sometimes users need a bit more hand-holding when trying things like this out for the first time, especially when your website is aimed at kids.
Overall i think we pulled it off really well. I think Katie herself is one of the best and most true representations of the human form that i’ve seen in Papervision3D to date. I think that the animations sync well even when changing clothes (which was no small feat) and that it performs very well even on a computer that’s not a super hot-rod. Now i just hope that kids enjoy plying with it. We had some young kids come in for the shoot for the demonstration and they seemed to love it, for me, that was one of the best parts of this whole job. Seeing people use the things that we make and having fun / finding them useful is one of the most rewarding parts of this job. Now go buy some dolls.
Flash Forward the TV show is about a mysterious global event that causes everyone on the planet to simultaneously lose consciousness for 137 seconds (2 minutes, 17 seconds), during which time people worldwide see what appear to be visions of their lives six months in the future. The event results in some deaths and leaves the survivors wondering if what they saw will happen.
The site that we have built is a hype-piece that shows a sneak-peak of the show. It’s basically a trailer viewer, but presented i a unique way using augmented reality
.
The instructions on the website inform the user that they must print this ‘marker’ and have an enabled webcam. Once you’ve got those two things you can start playing around with augmented reality. The show features a busted up city (which is the result of everyone falling unconscious for 1:17), and the marker that you are holding displays this city on top of it using some image recognition technology developed by our friends at Boffswana. The city itself was modelled in Maya and then brought into 3DSMax to be exported as a .dae file using a Collada exporter. We then use Papervision3D (via Flash) to import the model and display it real-time in 3D.
The image recognition works out which orientation the marker is on and so any direction it is facing in will always trigger the city to be directly facing the camera so that you can see it in it’s best view. Some of the nice little touches that we put into this city are promo billboards for other Channel 7 shows, a chinook circling the main building in the centre of the scene and a tiny video playing on a mini video screen on the main building.
You can twist the marker around in front of your camera to explore the city from different angles, but if you turn the marker around 180 degrees the back-drop of the city is revealed as a bigger screen with which to show the feature promo for the show.
It was a great project to work on with an extremely quick turn-around and a lot of technology flip-flops along the way. We were originally going to be using some hardware-based 3D technology that runs within a dedicated plug-in to show the city which would have allowed us to have a much more detailed city display and would have also allowed for natural feature tracking to be employed in stead of the high contrast glyph. This basically means that we were going to use a photo instead of the black and which marker to instigate the 3d scene. The spec ended up changing and we used flash with Papervision3D in the end. Which meant that we potentially had a wider audience using the site as there are no 3rd party plugins to install, but it also meant that we needed to re-model the city to fit within the confines of Papervision3D’s modest maximum polygon count.
At the end of the day i’m very happy with the result. Some may see it as a bit of a gimmick, and that’s fine, i respect that. in a sense it is, but it’s still damn fun to play with.
Just got a bunch of washers to use as weights for the slat of wood at the base. I’ve picked up about 38 of them, which worked out at about $15 including the bolts. This has been the most expensive part of the whole thing so far. All the PVC only cost me about $8.
.
