VFXhack

A terra cotta bunny, a playboy centerfold and a Mandril. What do these things have in common you ask? If your answer was the cast list to the greatest YouTube video of all time you’d be wrong (that would be cool though). The correct answer is that these are all images used and re-used in the testing process of creating digital imagery. The folks over the unfortunately named Art Fag City have put together a post of 20 CG Image Archetypes for your browsing pleasure. Lots of good history and links here. Take a look.

[LINK] (http://www.artfagcity.com/2008/07/28/img-mgmt-20-archetypes/) to the original post (marginally NSFW because it has a naked teapot in it)

Thanks to Azathoth for the link

The video proves once and for all that CGI and Radiohead are two great tastes that taste great together. The basic idea is this, screw the cameras 86 the lights scan everything and make the whole piece out of realtime streaming data. The result is imagery so deep in the uncanny valley that it’s cool. This behind the scenes gives a great overview of the technical and creative issues involved. I love the image of Lidar scanner on a dolly. The only problem, “Ready, and, SCAN!” just doesn’t have the right ring to it.

LINK to the full video thanks to JC for the link

What is “Nuking The Fridge”, you ask? Well, according to Urban Dictionary it’s…

A colloquialism used to delineate the precise moment at which a cinematic franchise has crossed over from remote plausibility to self parodying absurdity, usually indicating a low point in the series from which it is unlikely to recover.

Guy 1: “Wow. Did you see the new Indy movie? What the hell was that? It was like I was having some kind of flu induced absurdist nightmare.”

Guy 2: “Yep… did or did not that series permanently Nuke the Fridge?”

“Nuke The Fridge” has official replaced “Jump The Shark” as the term du jour on many Blogs and websites as well as IN the vernaculars of many VFX professionals. Does it trouble anyone else out there that a phrase previously used to conjure nostalgic memories of 70’s era schmaltz and lameness (c’mon Fonz, who wears a leather jacket while they are water skiing?) is now linked to an overwrought effects sequence? I’ve been bemoaning the fact that the modern VFX world is a lot more coulda’ and not enough shoulda’ on this blog for almost a year now and there are signs that things are coming to a head. Are Indy’s CG gophers and the technicolor dragon vomit that made up Speed Racer eliciting a collective cry of “Enough with the CGI already!” from the general public?

A case that a shift in popular opinion might be made based on the recent spat of blog posts and movie reviews based on the subject of a public grown desensitized and weary of movie magic. Scott Nye postED this tidbit on chud.com not to long ago.

But much as I support CGI, I do acknowledge it’s also an overused, lazy way to create what could have been an amazing sequence if done practically

And it’s not just the Matrix Generation thats bent out of shape about the state of visual effects. Johanna Schneller, movie critic at globeandmail.com observes…

I realize that CGI is a thrilling toy, a technology that seems to advance by the minute, that gets more fun the more you play with it. And I realize that in Hollywood, it’s become a tautology: Blockbusters have CGI, so if you want your film to be a bigger blockbuster, it must have ever-more-massive chunks of CGI. But I’m not sure that filmmakers realize what a yawn it is to sit through

This simmering animosity towards VFX has also jumped from film critics and nerds to the video game community. In a post on PTD magazine, William Stapleton asks these tough questions after taking his son to see the new Hulk flick.

what happens when the ‘gee-whiz’ wears off? My grandchildren will take CGI technology for granted, because they’ll grow up in a time when it’s the norm, instead of something that’s new and exciting. I’ll admit, I’ve gone to a few films where the only redeeming quality was the power of its special effects. But what will happen when special effects aren’t ’special’ anymore?

I realize that three fairly obscure blog posts do mot a backlash make, but I have to ask. Is this the begging of the end?. Visual Effects films, like any other cinematic style, have always been subject to cycles of popularity. The cold war fueled sci-fi boom of the 50’s eventually trailed off and didn’t pick back up again until the 70’s. It could be argued that we are on the downside of peak in interest in science fiction (the key driver of the VFX film) that was spurred on by the rise of the Internet. On top of that, we are quickly coming to a point where we are simply going to run out of comic books and sci-fi classics to “re-invent”. What’s next Power Pack:The Movie? Last Starfighter 2: Electric Boogaloo??

We all need to face up to the fact that the current trend of CGI blockbusters may come to an end some day. Heck even the sun will run out of fuel and burn itself out at some point. To survive VFX has to move past the gimmick phase and become a truly integral tool in production. Some films already do a great job of this (Children of Men comes to mind) while most use the technology as a crutch to prop up aging action heros or a way to create asinine comic relief for bad screenplays. Don’t get me wrong the computer artistry that goes into a lot of these films is astounding. Too bad it’s often wasted on a disappointing, or worse yet forgettable, piece of entertainment. Let’s hope that the entire VFX industry hasn’t collectively jumped the shark, I mean, nuked the fridge.

Thanks to VFXPlanet for the links to the stories used in this post

Just finished this little write-up for the work that I did for the series finale of Jericho. The whole show was a blast to work on and many talented people poured a lot of hours and hard work into this baby. Hope you enjoy.

The Mission

Jericho tells story of small town struggling to survive in the landscape of a post apocalyptic America. In the final episode of the show’s second season, the creators wanted to give the fans something special. The VFX team worked with production to give the audience a glimpse at what the rebuilt capitol of drastically altered capital city would look like and a stunning climax that pitted the shows main character, Jake against the might of the Cheyenne air force in a dramatic aerial battle. The challenge to the VFX team was to make these important story points completely realistic. The VFX was not to detract from the story but enhance it by delivering seamless effects.

The Approach

The reveal of the capitol building under construction in the new Capitol of Cheyenne was not to be your run of the mill matte paining. The story called for a city under drastic expansion and rapid construction. Reference photography of the current city of Cheyenne could be used as a base, but new buildings as well as construction cranes, vehicles, pedestrians and construction workers would have to be added after the fact. The show’s most difficult sequence came in the form of the final aerial battle. Knowing that it would be impossible to tell the story through the use of stock or custom shot aerial photography, the visual effects team settled on a mostly CG approach. The scene called for a variety of different aircraft and many camera angles to cover the action. There was dialog between Jake and the pilots of both the Cheyenne and later the Texas planes, so plates of the actors would need to shot on green screen and combined with the CG planes and backgrounds. At one point in the battle, two of the planes are blown out of the sky by air-to-air missiles. Knowing that a photo-real explosion would be difficult to achieve by using CG alone, it was decided that scale models were to be built based on the CG planes and animatics and blown up using traditional special effects methods. Later these elements would be composited with the CG planes.

Shot Elements

Once the methodology was decided upon, the wheels were set in motion to gather all the practical elements that needed to be shot. These fell into several categories. First HDRI 360° photography of the first unit location for the reveal of the Cheyenne Capital. The HDRIs captured on set were used as a basis for the background buildings and were used as a lighting basis for the 3d elements in the matte painting. The raw photos for the panoramas composed of 38 different camera positions with nine exposures for each position.

The final shot was to be from the perspective of a hotel room 12 stories above the street so the HDRIs were taken from atop a 40 foot platform. Many elements were also gathered to create the feeling of activity in the scene. A camera was again placed on a platform to match the high vantage point of the final shot. Many passes of ground activities of civilians, military, construction workers and their vehicles were taken. To create the illusion of more extras costumes were changed multiple times and movement patterns were varied from take to take.

The aerial battle plates consisted of two main parts, the green screen photography of the actors for inside the plane cockpits and the explosion plates of the jets. The green screens were filmed on a high definition video camera with Jake (the main character) sitting inside a cut out portion of a real plane cockpit that was a close match to the model that was shot on first unit.

The glass on the cockpit windows was left in on most shots so any of the actors reflections could be lifted and used in the final composite. The pilots of the F-15 and F-16 fighter jets were also filmed against green with the intention of modeling a digital cockpit for those shots. Pivotal to the success of the green screen shoot was the animatic that was created by the VFX team and cut by Jericho editorial. This animatic was worked on with the creators of the show so that all the major action of the scene was represented in a rough 3d form. This gave the actors an excellent idea of what was happening in the scene and helped them achieve the proper eye-lines and emotional impact rather than just acting against a void.

When it came to shooting the plane explosions, great care was put into the planning and preparation of the shoot. The animatic was again used to determine the detail and placement of the plane models. A VFX shoot team was assembled consisting of a DP and his crew, a pyrotechnics team and few model makers and painters. The plane that was to be destroyed was an F-15 fighter and the fiberglass shell from a remote control plane kit was used as the base model. Four shells with a wingspan of a little over 5 feet were prepped and painted to match the markings and textures of the CG models exactly. On the day of the shoot the model planes were packed with explosive and lifted onto a 30 foot armature. Three cameras were placed at different heights that lined up to several shots in the animatic, a live switcher was used to line up the video feeds from the camera to the animatics. The three cameras rolled simultaneously on each explosion at 120 fps. The higher frame rate was used to compensate for the smaller scale of the models. All together 3 models were blown up for the shoot.

3d Techniques

Before any modeling or texturing of the 3d assets was started, an exhausting search for reference was begun. The new Capitol building under construction in Cheyenne began as model from a stock library. Pieces of the substructure and framing of the building were then created using the stock model as a guide. Then pieces of the original model were stripped away to reveal the unfinished skeleton. The construction cranes for the 3d matte painting were modeled using photo reference, along with other street props. The hero buildings were generic models that were placed in the scene with projected textures of the building matte paintings used to enhance the details.

The aerial battle sequence started with the creation of a detailed animatic which served as a master template for animation, camera movement and lighting. After editing, the stand-in planes were replaced with hi-resolution versions. Careful attention was taken to make sure that the planes were authentic. The Cessna Citation that Jake flies in the scene not only had to look real in the CG shots but be a perfect match to an aircraft shot by production taxiing on the runway in the previous scenes. Many photographs were taken of the markings and logos on the plane to make sure the match was seamless. The F-15s and F-16s also had to be extremely detailed as they were seen close-up from nearly every angle. The cockpit and pilot seat of each plane also had to be rendered and textured to accommodate the green screen footage of the pilots. A lot of photographic reference both still and moving was used ensure that the shading and behavior of light across the planes surface was highly realistic. The flags and markings of the new Air Force in the Jericho story line also had to be created and applied to the planes.

2d Techniques

Many of the compositing shots in this show made use of the tight integration between the 3d and 2d working environments. For the Cheyenne Capitol shot, a 3d track was created for the scene. 3d elements were rendered with basic lighting using the track data, then the same track data was used in the compositing software. This allowed the compositor to place elements in the scene on 3d cards and have them track perfectly with the rendered elements. It gave the artists the flexibility to paint detail directly onto the parts of the image that needed it without having to request another element to be rendered in 3d. This technique was used to great effect when building the sky environment for the aerial battle scene. The shot began with the camera created during the 3d animation process. A basic sky dome and rough 3d landscape was added to each shot and rendered. The compositors than painted over the rough landscape using stock aerial photography to add detail while maintaining the basic lighting and depth from the 3d render. The 3d camera move was also brought into compositing for the addition of photographic cloud elements placed at different distances from the camera to create the illusion of volume and depth.

To create the photographic look of the planes, a multi-pass pipeline was used. Each component of the 3d render (reflection, color, specularity, etc.) was rendered as a separate element and combined in composite to achieve the final look. Using this method, the compositors had an extreme amount of control in how the subtle lighting effects played across the surface of the ships. Perhaps the most convincing of composite effects done for the show was the explosion of the two F-15s. Since the line-up of the shot footage of the planes exploding from the VFX shoot lined up perfectly, a quick dissolve from the CG plane to model footage was all that was needed. In fact the explosion proved so effective that close-up shots of the explosion from the VFX shoot were added to the edit towards the end of the editing process. The clean-up of the explosions involved removing the rig from each shot and pulling keys from the sky behind the model.

[ Javascript required to view QuickTime movie, please turn it on and refresh this page ]

Conclusion

It is rare to get to work on a Visual Effects sequence for a television show that i so pivotal to the storytelling of the episode. In the case of Jericho Episode 207, the story and visual effects were intertwined in such a way that one could not have happened without the other. Because of the tight integration between story and effects, all of the work done on the show had to be seamless. Anything out of place would distract the viewer from the emotional impact of the action. An imaginative combination of techniques both old and new were used to create scenes with an amazing degree of authenticity. The VFX team for this show worked tirelessly to make sure that the fans of the Jericho series would not be disappointed.

The Finns who brought you the movie Star Wreck have just released a teaser trailer for their new film Iron Sky. The basic plot is that Ratzis escape o the moon circa 1945 and are coming back to invade the Earth in 2018. The filmmakers plan to continue the theme of “mocking totalitarianism” is this release but what is more interesting to me than the actual content of the film is the way it is being made. Wreck was made using a large number of volunteers getting individual shots done in locations all over the world. Call it distributed filmmaking if you will. This endeavor seems to take the concept a step or two further and has set-up a collaborative website were users can sign onto the production and pitch shots and ideas for scenes. It will be interesting to see how a single Directors creative vision meshes with a social networking style approach to film production.

Link to the Wreck A Movie collaborative filmmaking platform

Nyarlathotep passed along this awesome pic that at first glance looks like some sort of Sci-Fi S&M setup but is really an early motion capture rig. There is more info on Dave Sieg’s Scanimation website about this image

I recall seeing a demonstration videotape in which a girl wore a harness equipped with motion sensors connected to a Scanimate. Each sensor was a goniometer, a type of rotary transformer which converts angular displacement into a corresponding phase shift between windings. The wearer of this harness could cause a stick-figure, generated by Scanimate, to mimic his movements in real time. Remember, this was done back in the early seventies, well before any of the recent magnetic or optical motion capture work.

I’ve searched the whole internet (well not really) trying to find the full demo video of this system but to no avail. Any hints from fellow VFXhackers out there could help us all scratch our vintage MoCap itch.

LINK to a cool movie of the Scanimation system in action for Dave’s site

This Bollywood style dream sequence takes you inside the mind of a CG animator faced with the prospect of life after outsourcing. It’s pretty funny but falls a bit into the category of gallows humor considering all the people in this video could be replaced someday soon by their Indian counter-parts. Ah well, if you can’t laugh at yourself who can you laugh at?

(via Cartoon Brew)

This clip is 10 minutes of pure VFX gold. It’s a behind the scenes explaining the process behind the Death Star mission briefing in A New Hope. There are some great details here and the process really is a combination of practical and computer techniques. I love that the animation wasn’t keyframed but “performed” real-time by manipulating a bunch of dials and knobs. The creator of this animation was Larry Cuba, a CG pioneer who programmed the film Arabesque with with John Whitney.

LINK to an excerpt of Cuba’s 1985 masterpiece Calculated Movements

Special thanks to Agrapha for the Star Wars link

AWN has posted an article revealing the results of an animation survey filled out by over 300 readers. It’s not surprising that most of the surveyed are guys who have a boner for hand drawn animation. It seems that they are particularly disturbed about the direction the winds of change are blowing in regards regards to mocap. Mark Simon sums up the issue thusly.

Sony Imageworks and Beowulf director Robert Zemeckis spent over $150 million producing a motion-capture CG animation movie. Want to know what scored the lowest in the survey of favorite animation styles? Motion-capture CG animation. Barely 2% of the survey-takers prefer this style of animation. It would seem that this is not a good investment.

I do have to agree that the Citizen Kane of mo-cap movies has yet to be made. And It looks like the movie biz has a long way to go in convincing the animators of the world that scaling the cliffs of the uncanny valley is worthy a pursuit.

LINK to full survey results (via VFXPlanet)

I’m always on the lookout for a new way to simultaneously do something productive and keep from getting bored while I’m on set waiting for a VFX shot to come up. My latest arts and crafts project involves creating the ultimate LED tracking marker. Now you may ask “Why LEDs? Why not a triangle inside a circle, or newspaper, or a tennis ball for god’s sake!” Well those methods are still valid but the in the right situation and LED tracking marker can really do the trick. Here are some reasons why you might want to go the LED route.

1. LED stands for Light Emitting Diode. As the name suggests you don’t need to light them because hey are a tiny light source onto themselves. This is perfect for areas of the frame that don’t have enough exposure for your common variety tracking marker.

2. LEDs hold up better when they are out of focus. When you shot a greenscreen with a long lens and an open F-stop, the background of your shot is gonna go way out of focus. Regular tracking markers turn to mush in this situation causing more a problem for clean-up than helping with the tracking process. The LED will bloom but tends to hold it’s intensity and not break-up as much.

3. The are small. LEDs are compact and light. A little bit of snot tape on the back and the can be affixed to anything from a patch of skin to a set wall. Ball up in some strong tape (sticky side out of course) and you can even throw them up on a wall that’s too high to reach. No more asking that grumpy grip if you can borrow his ladder.

4. Power is not an issue. A single battery can keep an LED running strong for days depending on the temperature. Battery life deteriorates rapidly in the cold so if you are shooting outside in the winter, make sure you grab a few packs of extra batteries for your kit bag.

There some things to watch out for with LEDs though. Their biggest strength (the fact that they emit light) is also their biggest weakness. LEDs can spill unwanted light all over the place if they are to close to (or on) the subject they are shooting. Fortunately, LEDs come in different flavors that include color variations and frosted plastic casings. The frosted casing cast a lot less spill but also need to be aligned to point at the lens for maximum effect. Try and find some samples and try out some color and casing combos before you head out. You never know how things could change on set, so it’s important to carry a bunch of LED options with you in your kit bag.

At this point you may be asking yourself “Wow, LEDs sound great. Where can I buy some?”. Well, the short answer is you can’t. You have to build LED tracking markers yourself. But never fear intrepid VFX explorer! Here’s a step by step on on how to create your very own set of LED tracking markers.

Step 1: Buy some LEDs

You can get these babies on-line or at your local electronics supply store. The come in different colors and you can also find them with frosted plastic. You can frost clear ones yourself with some dulling spray or spray mount. You can also file off the top of the plastic to expose the diode if you want them to be even brighter.

Step 2: Buy Some Heat Shrink Tubing

This is a great trick. Heat shrink tubing is a strip of vinyl that shrinks down to 1/2 it’s original size when exposed to heat. You can get it in all different kinds of colors including green which makes it almost invisible when affixed to green screen.

Step 3: Buy Some 3 Volt Batteries

These are the kinds of batteries that most cameras use. You can buy them from a store but it’s much cheeper to buy them in bulk on-lne.

Step 4: Cut the Tubing, Slide in the Battery Apply Heat

Cut the tubing so that is just a little bit larger than the battery, then just slip the battery inside the tubing. Apply heat with a heat gun and “Viola!”, the tubing makes a nice tight jacket around the battery. Make sure when you are applying heat that you hold the battery and tubing with a pair of pliers so you don’t burn yourself.

Step 5: Insert LED

All that’s left to do now is slip the wires of the LED between the battery and jacket and you’ve got yourself a tracking marker. You can stick these to surfaces using double sided tape, stick them to people with a Band-Aid or just rest them on flat surface. Happy shooting!

LINK to a del.icio.us tag with places to buy the stuff you need to make your own.