BitTorrent usage increases in Europe, following the blockade of The Pirate Bay

In a twist that will surprise no one except the RIAA, MPAA, BREIN, and other anti-piracy lobbies, the amount of BitTorrent traffic has stayed the same or increased in Europe followingthe blockade of The Pirate Bay in the UK, Netherlands, and other countries.

This news comes from XS4All, one of the largest European ISPs, which haspublished a graph of the network traffic associated with the BitTorrent protocol (pictured below). The left side of the graph is January 2012, the right side is June 2012, and the red line signifies February, when Dutch ISPs were ordered to block The Pirate Bay. While it’s hard to make a qualified decision without seeing data from 2011, it definitely seems like traffic hasn’t decreased — and might have even increased slightly. This data aligns with research from the University of Amsterdam, which also found that the Dutch Pirate Bay blockade had no effect on the total amount of BitTorrent traffic.

This data strongly conflicts with BREIN — the Dutch anti-piracy lobby — which, just a few days ago, announced with much jubilation that The Pirate Bay is now unreachable by 90% of Dutch users. How can there be such a disparity? There are three likely explanations: a) The UK and Dutch blockades created a lot of publicity (and no publicity is bad publicity); b) The Pirate Bay isn’t the only torrent site, and most torrents are available from multiple sites; and c) Veteran internet users are a lot more savvy than the RIAA, MPAA, and BREIN give them credit for — it’s awfully easy tocircumvent the blockade with a proxy or VPN.

Yet again, this is strong proof that trying to force a change in behavior simply doesn’t work — especially on the internet, where its denizens value freedom above all else. History is full of shutdowns and blockades — Napster, Kazaa, Limewire, Megaupload — and yet file sharing is still just as prevalent. Instead of pissing away billions of dollars on lobbying governments and law enforcement agencies, the only way to truly curb file sharing is to provide an equivalent or better service than torrent sites and digital file lockers. Steam and iTunes have proven that consumers want to pay for their multimedia fix.

The sad truth, though, is that it’s actually easier to persecute file sharers than to navigate the insane morass of rights holders and licensees on any given TV show, movie, game, or song. If it was easy to create an official, legal version of The Pirate Bay, then the entertainment industry would’ve done it already — they’re not that stupid. After years and years of evidence that criminalizing file sharing doesn’t improve matters, though, you think that the MPAA and co would invest their efforts elsewhere, though — such as hammering out their own Pirate Bay, or creating albums and movies that don’t suck.

Read more about piracy

Source

How digital technology is reinventing cinema

As of 2012, around 60% of all cinema screens worldwide have been converted from film to digital projectors. Over half of those digital screens are outfitted with stereoscopic (3D) projectors. By 2015, it is expected that almost every cinema screen will be digital and that film projection will all but die out.

Digital cinema is a lot more than just a digital projector, however — the film industry, as I’m sure you’re aware, is a multi-billion-dollar behemoth, and digital cinema is probably the biggest shake-up since the advent of 35mm film itself. There’s a whole framework behind digital cinema, from filming, to digitization, to distribution and projection, with some seriously cool tech along the way — which, of course, we’re going to dig into.

Filming

For the most part, movies are still predominantly shot using 35mm film stock. Cinematography is certainly moving towards digital cameras, but the legacy of film is so great — the equipment, the process, the human expertise — that it won’t disappear for a long time. With the rest of the movie making process being almost entirely digital — from editing, to distribution, to exhibition — digital footage is a lot easier and quicker to work with. Just as digital photography usurped film photography, digital video cameras are destined to replace film video cameras, with digital cameras from the likes of Canon and Red leading the way.

Digital intermediate

The irony of using film cameras, though, is that they’re all scanned into a digital intermediateanyway. Almost every big film of the 2000s was converted from film to a 2K (~2048×1080) digital intermediate — so even if you think that film has a higher resolution than 2K, or if the grain is somehow more attractive than pixels, tough luck. If the film is shot with a digital camera, then this scanning stage (which is quite expensive) can be skipped.

Once you have a digital intermediate, talented artists take care of the editing, color grading, and CGI (computer generated imagery).

Distribution

In 40% of cases (conventional projection cinema screens), the digital intermediate is then transferred back onto film, and copies are made (at a cost of thousands of dollars each) for each cinema that will be screening the movie. For digital screening, the digital intermediate is exported as adigital master, which includes all of the video, sound, and data required to project the movie correctly.

Now we get onto the techie bit of digital cinema. Before distribution to cinemas, the digital master is encrypted and compressed into a Digital Cinema Package (DCP), which is a standard format defined by Digital Cinema Initiatives (a joint venture by the major movie studios).

A DCP contains a bunch of multi-gigabyte MXF (Material eXchange Format) files and playlist/index XML files (very similar to the way a DVD contains VOB and IFO files). MXF stores video in JPEG 2000 format (an updated version of JPEG), at 2K resolution up to 60 fps, 4K resolution up to 30 fps, and 2K 3D at 48 fps. XYZ color space is used, with 12 bits per pixel precision (36-bit color). The max bitrate of MXF video files is 250Mbps, or around 30MB/sec. This means a single movie in DCP format is around 200GB. (By comparison, Blu-ray movies generally have a bitrate of around 30Mbps, or 3.8MB/sec.)

The audio MXF files use a standard WAV container with a 24-bit sampling rate of 48 or 96KHz. Up to 12 separate, concurrent audio channels can be used.

Most importantly, these MXF files are encrypted using 128-bit AES — if they fall into the wrong hands (those of a pirate, say) it is virtually impossible to decrypt them.

Finally, the DCP is copied onto a hard drive, which is protected by a rugged enclosure (usually a CRU DX115). These hard drives are then distributed to cinemas via courier. It is also possible to deliver the DCP directly to the cinema via high-speed internet connections, though this isn’t usually done.

Digital cinema servers

Once The Avengers DCP hard drive arrives at the cinema, it’s slotted into a digital cinema server (DCS). A DCS is basically a proprietary, rack-mounted computer that has a hot-swappable hard drive bay on the front, and a bunch of video/audio outputs on the back. There are many companies that produce DCSes, but Dolby (pictured below) and Sony are two of the frontrunners (and their systems run Linux!)

When a new hard drive is slotted into the DCS, the movie is “ingested” onto local storage (usually a big, multi-terabyte RAID). This is where the encryption kicks in: Every DCP hard drive is encrypted with a key that is only known by the target DCS. In other words, whoever originally encoded the DCP has the cinema’s public key — and then when the movie is ingested, the private key is used to decrypt it for playback. The certificate that accompanies every DCP also defines how long the movie is valid for (i.e. the studio can force The Avengers to expire within a month, if it wants). This ensures that DCP copies don’t go walkabout, and explains why the pirate movie scene still has to rely on cam, telesync, and telecine releases (though I wonder why no one has managed to extract that private key from a DCS yet…)

It is possible for a cinema to have just one central DCS that outputs to every screen. I’m told that a single copy of a movie sitting on a DCS can be streamed to more than five screens at the same time (though with each stream requiring 30MB/sec, it’s more likely that we’re talking about a cluster of DCSes all slotted into the same rack, each with a mirrored copy of the film). Generally, a cinema’s DCS is controlled using a Theater Management System, which could be a discrete computer or just a web-based interface running on the DCS. The TMS allows employees to set up a playlist for each of the movies currently showing at the cinema.

Digital projection

And finally we arrive at the bit that most consumers actually care about: digital projection. Because most digital intermediates over the last decade have been at 2K (2048×1080) resolution, a lot of cinemas are still outfitted with 2K projectors. With the emergence of 4K digital cameras from companies such as Red, and 4K projectors from the likes of Christie and Sony, 4K digital cinema is making inroads.

Digital projectors are beastly machines, with huge, 4-kilowatt xenon lamps that produce tens of thousands of lumens. Price-wise, these projectors — such as the Sony SRX R320, pictured right — are usually around 50,000 to $100,000, with replacement lamps costing more than $1,000 each. Fortunately, for that price you do get 3D projection thrown in for free. At the core of nearly all digital projectors are Digital Light Processing (DLP) chips made by Texas Instruments.

In case you’re wondering, these cinema projectors are worlds apart from the Sony and Red 4Kprojectors that we’ve written about recently (and they’re significantly cheaper, too). The Sony and Red models are home theater projectors, with lumen ratings of a few thousand — about a tenth of the Sony SRX R320, which needs tens of thousands of lumens to illuminate 20+ meter screens.

The future of digital cinema

There is no doubt that cinema is moving away from film and towards digital — but is that a good thing? Digital distribution is undoubtedly more flexible than moving around bulky, costly film reels — but on the flip side, what about archiving? Properly-stored film stock can be archived for a hundred years, and unlike hard drives and other digital storage mediums, there’s no risk of film playback ever being antiquated (all you need is a lamp).

Driving down the cost of distribution has also significantly empowered student and independent film makers, and as the cost of digital cameras and storage continues to drop, this trend will only accelerate. Much in the same way that we’re drowning in digital photos, we will soon drown in digital videos and movies (if we’re not already).

Looking forward, the resolution of digital cinema will eventually hit 8K (which should finally quiet any digital image quality naysayers), and we should begin to see a lot more movies played back at higher frame rates like 48 and 60 fps. The Hobbit will be the first worldwide release of a 48 fps film, shot entirely with digital Red 4K cameras — and I’m sure it won’t be the last.

Source

Why movies are moving from 24 to 48 fps

Just as talkies replaced the silent film, and color has become the norm for feature films, avant garde, influential directors including Peter Jackson and James Cameron are determined to make high-frame-rate films, along with 3D, the next big thing in movies. Their crusade took a bit of a step sideways recently when the screening of clips from Jackson’s new film The Hobbit at CinemaCon was met with decidedly mixed reviews. Like it or not, though, expect to see more theaters equipped to show high-frame-rate movies, from 48 frames per second all the way up to 120 fps, and an increasing number of movies shot at higher frame rates. Jackson himself was stubbornly defensive after the screening, with PC Magazine characterizing his response to criticism as “deal with it.”

In technology, where more is almost always better, it is unusual to hear complaints about something getting faster. Some viewers of the preview complained that the scenes were “too real” — the 48 fps recording rendering the action so lifelike that they had trouble remembering that Middle Earth was supposed to be a fantasy. Other viewers missed the languid feel of traditional 24 fps movies. Unconsciously, we’re also used to 30 fps and above content being “made for TV,” and are accustomed to feature films showing at the lower 24 fps rate.

Film look sacrificed for the sake of 3D?

3D is one of the driving forces for high-frame-rate movies. By shooting at 48 fps, it’s possible to show 24 fps to each eye through a pair of active glasses, for example. It’s probably no small coincidence that Cameron and Jackson are two of the largest promoters of 3D movies, not only shooting them that way but with Cameron converting his own Titanic to 3D for re-release. Like 3D, proponents of high-frame-rate argue that it just takes time to get used to the more realistic images. Jackson says that once audiences see an entire feature film like his The Hobbit at 48fps, especially in 3D, they’ll learn to like it. Perhaps fearing more audience backlash, Jackson has decided not to release any 48 fps trailers for the film — saving the faster frame rate for its debut in theaters on December 14.

48 fps also allows for the creation of very smooth slow-motion scenes, simply by double-printing each frame to yield a 24 fps half-speed version. Of course in this case 48 fps could be used just for the scenes which need to be in slo-mo, with the rest of the film recorded in 24 fps. Even for full-speed scenes, 48 fps has advantages. Fast camera moves no longer cause “strobing,” and individual frames are sharper. Action scenes are definitely smoother and more lifelike. These changes may be disconcerting to those used to viewing movies at 24 fps, but new moviegoers could quickly become addicted and not want to go back. Just like the rush to color led to a flurry of colorized versions of black and white classics, we may well see post-production 48 fps renderings of existing movies.

The business of 48 fps

There is another big reason studios and theater chains will be pushing 3D — money. With increasingly capable home theater setups and Internet streaming competing with theaters, the movie business needs to differentiate its offerings in any way it can. Upgrading theater projectors to 48 fps, even at a cost of several thousand dollars per screen, may pay for itself if it gives theater goers a premium experience. Jackson is hoping that over 10,000 theaters will be high-frame-rate capable by the time The Hobbit releases. Even so, Jackson and Warner Brothers are hedging their bets — the film will be released in six different versions: 2D, 3D, and 3D IMAX — all of them in both 24 fps and 48 fps.

High-frame-rate has attracted the attention of camera makers also. Previously confined to specialty, high-end, camera makers like Red, Canon has added support for frame rates up to 60 fps to itsupcoming Canon EOS C500 – although the jump from 48 fps to 60 fps is hard to detect, so most high-frame-rate movies will likely settle for 48 fps. While still pricey, high-frame-rate tools will increasingly become cost effective even for independent filmmakers wanting to take advantage of the new technology. Even with some well-known filmmakers like Ang Lee being skeptical, the big money and big names behind 48 fps movies mean it’s here to stay, so we’ll all have to hope we can get used to it.

Source

Nanocopters, Smartbirds, and beyond: The coolest robot videos ever

Hardly a week goes by without an amazing new robot video showing up on the web. After diligently watching well over a hundred of them, I’ve collected a number of videos that demonstrate the incredible capabilities of modern robotics. Think of this as a primer on awesome robot videos as well as a refresher on some of the most viral technology videos to hit the web recently.

Have fun watching, and let us know if I’ve missed your favorite.

Best of the nanocopters

Quadcopters are definitely cool, like the construction ‘copters we wrote about earlier this year. Nanocopters, though, bring awesomeness to a new level, with their maneuverability and dare I say cuteness. As robots go, they aren’t the smartest on their own — since they rely on nearby computers for high-level programming and in this case their vision — but as a system they are capable of some amazing stunts. These two quadcopters, shown in the ETH’s Flying Machine Arena in Zurich, can respond to the ball more quickly than a human pilot would be able to.

To give you a better idea of what swarms of nanocopters are capable of, here is a video from the University of Pennsylvania’s GRASP lab, featuring a variety of solo and “swarm” stunts.

However, if you think humans are about to let a bunch of plastic ‘bots take the formation crown, think again. These Japanese show that humans can perform in unison as well:

If you’d like to learn more about what it takes to program a swarm of nanocopters, there is an excellent TED Talk on the how to of building and programming nanocopters:

Finally, if you haven’t seen it yet, robots indeed have their own version of silly kitty videos. Here the University of Pennsylvania’s swarm is programmed to play the theme from James Bond:

 

Boston Dynamics’ Big Dog

Our tax dollars pay for a lot of different things, but one of the coolest is the DARPA-funded Big Dog robot from Boston Dynamics. The first robot capable of walking across rough terrain, Big Dog and its relative, AlphaDog still hold the crown for best mechanical pack animals.

Robots like Big Dog and others funded by DARPA will help lay the groundwork for the newly announced DARPA challenge to build a humanoid rescue robot.

 

Haile, featured in this video, doesn’t look all that human, but if you close your eyes, you can definitely imagine that it is a real — if novice — jazz drummer. To get behind the scenes and look at what it takes to build a robot that can not only play the drums but also jam along with other musicians, Georgia Tech provided this video of its work in creating Haile, a robotic percussionist:

 

DIY amusement park ride in Google car

Who needs an amusement park when you can program your self-driving car to race around a parking garage at high speed like these Googlers. Imagine the day when we can all program our own fun — hopefully not all at the same time. For those who may worry about the high-speed thrill ride in the video, I’ve passed Google’s cars on the road where they meticulously cling to the speed limit — and, so far, the right lane.

 

SmartBird

As fun as it is to watch nanocopters, they’re pretty simple to comprehend. A few little motorized rotors, a battery, a gyroscope, and a spiffy control system. Flying using wings is an entirely different challenge. Even the physics of modeling a bird’s wing movement is complex.

Festo’s SmartBird does an amazing job of modeling the flight of a herring gull — using only its wings to take off, fly, and land. Ultralight construction helps make this involved form of flight not only possible, but elegant. In fact, if you didn’t know this was an article about robots, it’d be easy to think we’d gone bonkers and were now posting random videos of flying birds. Think I’m kidding? Here’s a video of gulls mobbing SmartBird, thinking it’s a threat:

There is a lot of fancy science behind SmartBird, which Markus Fischer, head of R&D for SmartBird maker Festo, explains it in this “how-to” video he presented at TED:

 

Towel-folding:

For decades people have speculated on robots helping us with housework. A brace of vacuum-cleaning robots have shown up, but most other tasks have proven beyond their reach. This UC Berkeley video of a towel-folding robot, as impressive as it is, shows one reason why — speed.

While the robot does eventually figure out how the towels — that it has never seen before — are arranged, the video needs to be sped up 50x to be watchable. A human can perform the same task almost effortlessly and quickly — although perhaps for some of us, not as neatly. As with most computer-based tasks, though, now that it can be done at all, it’s only a matter of time before a robot is as fast as a human.

 

Asimo just keeps getting better

No list of robot videos would be complete without at least one featuring Asimo, the venerableshowcase robot that has been around for over a decade. Each year it learns some new tricks. In this video Asimo shows that it can walk around people, keeping its balance while changing course. However, it can still only do this task we take for granted slowly and carefully — once again reminding us just how amazing humans actually are.

 

Honorable mentions

Self-assembling robot research funded by the EU.

A Rubik’s-Cube-solving robot.

And finally, the cat-petting, Wii-and-Kinect-based robot ExtremeTech covered in December.

That’s all the robotics awesomeness for now. If there is anything the article missed, please leave a suggestion below.

[Image Credits: SmartBird]

Source

Titanic: How do you convert a movie to 3D, anyway?

Not everyone likes 3D movies, and almost everyone agrees that traditional movies later converted to 3D have been disappointing. Until now, James Cameron, famed director and now also deep sea explorer, agreed with them. He has been a harsh critic of 3D film conversions, except that he has now spent a reported $18 million converting his blockbuster movie, Titantic, to 3D. The re-release is timed to coincide with the movie’s 15th anniversary and the 100th anniversary of the Titanic’s sinking. This herculean effort inspired us to take a closer look at how a movie gets converted to 3D after it has already been shot, and what makes Cameron’s effort so daunting.

Before a director — we’ll use the term director to refer to whichever filmmaker has taken on the task of spearheading the conversion — gets into the technical process of starting the conversion, some decisions need to be made about the look and feel of the converted film. Most importantly, the director needs to decide how far from the audience the action and sets are to be. Monsters or bullets appearing to fly out of the screen towards the audience add to the “wow” factor of a movie, but can cause nausea in some moviegoers and may just not be appealing to others. Alternatively, the depth provided by the 3D rendering can be used to show perspective and distance, pushing the background of the scene further away from the audience. Different scenes can combine these approaches, but varying the technique too much leads to motion-weary viewers.

Managing depth with parallax

The key to making a 3D photo or video appear to have depth is parallax. Simply put, parallax describes the way an object’s position or direction changes, depending on viewing angle. We all know that when we cover one eye or the other we see slightly different views from each eye. That makes filming a 3D movie deceptively simple — just place two cameras the same distance apart as our eyes, and let them rip. In reality it is much more complex than that, but at least the director has both views to start with. Converting a 2D movie to 3D requires synthesizing those two views for each of the well over 100,000 frames of a feature-length film.

When creating depth using parallax, the director can use either negative parallax, which makes objects appear closer than the viewing screen, or positive parallax, which makes them appear further away. It’s easier to explain with an illustration, like the one we’ve provided here.

Positive parallax is limited by the “artificial horizon” — the parallax of a point at infinity. The artificial horizon is rendered simply by having objects on the screen appear as far apart as our eyes, about 2.5 inches (6.4cm). For 3D TV that may mean that the left and right images would each be shifted quite a bit to make an object appear at infinity. Unfortunately most people don’t do very well at fusing images that vary by more than about 2.5%, so it isn’t possible to provide unlimited depth cues on a screen smaller than about 100 inches (2.5m) wide — at 100 inches the 2.5 inch desired effect is just about 2.5% of the total screen width. While 100-inch-wide screens are out of reach for most 3DTV viewers, the same effect in a movie theater requires a much smaller relative shift in the images, typically 1% or less.

Negative parallax — objects popping out of the screen — is also limited by our ability to fuse images quickly. Even more than positive parallax, overuse can result in viewer discomfort. A good reason not to sit in the front row of a 3D action flick if your eyes aren’t up to the challenge. Good 3D conversions are designed with the front row in mind, with the 3D “wow” effects fading as you sit further back in the theater.

When a movie is shot in 3D, the scenes can be planned so that the parallax doesn’t change quickly from scene to scene — making it easier for you to follow the action and for their eyes to adjust to each scene. Converting a movie which has already been shot doesn’t allow for that luxury, so in addition to all the other challenges of planning for the 3D conversion, smoothly managing transitions for comfortable viewing is also a necessity.

Adding depth to a 2D movie

Once the director has decided on the ideal depth range for each scene, then the technicians can get to work making it happen. First they need to establish a depth map (pictured below) for each frame. The depth map is essentially another layer or channel overlaid on the image that indicates the depth of each pixel from the viewer. The simplest possible way to estimate depth is by assuming bright objects are closer than dark objects, as is done by some of the desktop software tools that can create pseudo-3D images from 2D originals automatically.

Fortunately for those converting feature films, they already know how far away each object on the set was — or was supposed to be — when the movie was made. Productions stills also provide additional information. While this means they don’t have to guess or estimate depth, humans still need to painstakingly assist the software in adding depth information to hundreds of thousands of frames. This is no doubt where many of the 450 people who worked on the Titanic 3D conversion for over a year spent their time.

Another plus is that post-production software is excellent at tracking objects through a scene once they’re defined, but human help is required to ensure each one is accurately positioned when it first appears. Cameron wasn’t immune to the seemingly interminable scope of the effort, having reviewed the movie frame by frame for the conversion and calling the process “mind-numbing, like mowing your lawn with a toenail clipper.”

Once depth maps are made for every frame of the movie, the technical process of creating a stereoscopic image of each frame can begin. Filmmakers use expensive special-purpose tools, but for everyday use the process is as simple as using the Displace filter in Photoshop or the “Convert to 3D” script in the G’MIC plug-in for GIMP. This screenshot (below) shows the G’MIC dialog for converting images or video to 3D, either automatically or when a pre-computed depth map is available. Unfortunately for anyone hoping to show the results in a theater for big bucks, the simple approach falls a bit short.

A major issue our simple approach doesn’t handle is a side effect of looking at a scene from two points of view. Each view sees slightly different objects, and pieces of objects. When converting a 2D film to 3D, that means objects which aren’t in the original scene suddenly appear, and others have additional surfaces revealed. They need to be filled in or estimated somehow during the conversion. Sophisticated conversions actually use multiple layers of depth maps — showing what is at each pixel for a whole series of depths. This allows the conversion software to reveal the appropriate “hidden” object or surface as the perspective changes. Even allowing for multiple layers in the scene doesn’t guarantee that the resulting image will appear natural. Part of why Cameron had to look at every frame more than once is the subjective way we all perceive reality and decide for ourselves what looks real.

Keeping the 3D real

Our subjective response to other depth cues further complicates the conversion process. As we watch a 3D movie, our eyes and brains are responding not just to the 3D effects provided by the director during the conversion, but to all of the depth cues we always use to assemble scenes in our mind. There are many “tricks” we use to do this, and they vary from person to person. One simple example is that we assume that if one object hides part of another, the hidden object is further away from us. Another is that if we have a linear perspective along the side of an object we assume it is receding into the distance (pictured below). We also assume that objects appear faded with distance, so hazier portions of a scene are assumed to be further away, and that objects which appear smaller are in general further away than ones which appear larger.

Parallax is actually one of the weaker of the ways most of us estimate the depth in a scene. You can verify that for yourself by closing one eye. You can still get a good sense of distance — as anyone with one eye can attest. When the depth cueing provided by a 3D conversion is at odds with any of the many other visual cues already in the movie, the 3D effect will be lessened and may result in viewer discomfort. There isn’t any automated way to guarantee that the parallax matches our other depth cues — thehuman visual system is too complex and not well-enough understood, at least not yet. Once again, frame by frame review, like the one done by Cameron, is the only way to guarantee the best quality. Cameron is certainly no stranger to persistence, as hissuccessful quest to dive to the bottom of the Mariana trench has shown.

Converting Titanic

One reason Cameron says that previous 3D conversion efforts have failed is that they were shoehorned into an already busy production schedule, meaning the filmmaker didn’t have the dedicated attention needed to ensure the conversion came out the way they wanted. He also credits his learnings from Avatar with helping him know how to work on this conversion — experience which most other filmmakers don’t have. By contrast to live action films, it’s worth pointing out that computer-generated movies (often called CGI) are relatively easy to convert, if all the original software and models are still available. Instead of simulating 3D from one camera position, another camera position can simply be added to the rendering software.

For Titanic’s 3D conversion, Cameron relied on relied on William Sherak and his team at Stereo D for the bulk of the effort. They started by having a team of over 300 artists rotoscope the original movie to outline each object, probably using a tool like Imagineer’s Mocha. This allowed them to build a 3D model of each scene to which they could add camera motion and create a depth map for each of the nearly 300,000 frames. The process relies on specialized and often highly proprietary tools such as those developed by the In-Three team, now part of Cameron-founded Digital Domain Productions.

It would be tempting to say that the more-than-two-year (over a year of planning and over a year of doing the conversion), 18-million-dollar project to convert Titanic to 3D has been a labor of love — except, of course, it’s a great way to generate more revenue from one of the top-selling films of all time. Cameron explained his desire to convert and re-release Titanic not by touting the benefits of seeing in 3D, but by merely recounting that he wanted “to put Titanic back on the big screen.”

[Image credits: Parallax, Depth Map, Shiva statue]

Source

Marco Tempest tells the story of magic at TED, with Kinect-powered augmented reality

Marco Tempest, the techno-illusionist who shot to internet stardom last year with his three-iPod/iPhone magic act, has followed up with a augmented reality, Kinect-powered TED talk on the history of storytelling and illusion. The six-minute video, dubbed “A magical tale,” is embedded below.

Like the iPod illusion (embedded at the bottom of the story), Tempest combines sleight of hand and digital computers to create a very satisfying performance — but this time he goes one step further and weaves 3D augmented reality into the mix. On the hardware-side, Tempest simply uses an HD video camera and Microsoft’s Kinect, but the software — bit that actually produces the illusion — is really quite complex.

Tempest, working with two software development houses, Onformative and Checksum5, used VVVV to script the six-minute magic trick. VVVV is a hybrid programming environment that allows you to mash together video clips, real-time special effects, and the input from external devices. In “A magical tale,” every special effect is rendered in real time using a desktop PC — the particle effects, the face tracking (and re-mapping), and digital compositing (such as the big, 3D globe that Tempest steps inside); they’re all reactive to Tempest’s movements. The end result is impressive, if not quite as smooth or magical as the iPod/iPhone trick. Embedded below is a “making of” video that illustrates the technology behind the illusion.

Ultimately, Tempest says that “A magic tale” is more about engaging an audience than bona fide illusion. “Magic is a unique tool for pre-visualization,” he says. Instead of a magic trick with sleight of hand, the subject matter could instead be a video game, an interactive and interesting way to teach kids otherwise dull subjects, or even an augmented reality face-to-face Skype chat. It won’t be long until Kinect-like sensors are built into laptops and smartphones, don’t forget.

Read more at Creative Applications, or hit up Marco Tempest’s website

Source