{ illuma·blogspot·com }

Sunday, January 30, 2011

••◊ Street Photography from the LA Fit Expo

First...I know, I know, I'm supposed to be doing part 2 of the frame rate conversion tutorial - which I'll get to this coming week. Today is a recovery day for me after an short adventure working at the Los Angeles Fitness Expo yesterday. Ironman hired me on for the day to work at the bodybuilding show. Of course, I took this as an opportunity for street photography since I'd never been to the LA convention center. Turns out that the place is lit up like Vegas. Large LED signs blaring out at all hours of the day (still going at 2am and 5am with the sirens); music pumping into the Nokia plaza across from Staples Center; couples dressed up out on the town for the evening. Plus I was surrounded by fitness industry professionals whose lifestyle and demeanor always makes for interesting photography.

The evening before the show I had a chance to walk around Staples Center and Nokia plaza. The LA Lakers were playing that night so the plaza was packed until about 8pm, then everything suddenly went calm.



For dinner I grabbed a seat at Wolfgang Puck's on the plaza so I could people watch. As you can see, I didn't have any issues with the meal.



What comes after dinner...breakfast! It's why I wake up in the morning. This particular morning I hit the Pantry Cafe, a legendary spot for breakfast in LA at a quarter to six. Lonnie recommended it the night before and I wasn't about to go searching blocks for breakfast. A boy's got to eat! At that time in the morning their customers are mainly the hard working folks looking for a solid meal to start the day. Very 50's vibe. The pancakes there were so good that I didn't even need syrup. From my colleagues comments; by the afternoon they have a waiting line going out door.



The fit expo for me was all about the bodybuilding show, which I was hired to sound engineer for the day. Now mom, before you thinking I'm all chauvinist and stuff I have a good reason there's only pictures of the figure girls. The bodybuilders require sound changes every two minutes so I have to work and pay attention to the show. For the figure/bikini competition I only have to put on some happy pop tunes and let it play - so I have time to play. That and...



Time for some gratuitous celebrity shots. I mean it's Los Angles, right. Can't have a show without one or two or three or Paris Hilton. At the fit expo I had the pleasure of working with Phil Collen (lead guitar/Def Leppard). He and his wife Helen are two of the most friendly and genuinely nice people you could hope to meet. They're starting a fitness business called Physical Mechanix, as you can see from the shirt. I still remember rocking out to "Poor Some Sugar On Me," despite being banned from watching MTV. Now I think about the openning time lapse of the stage assembly as well. Ah...the days of being a Charvel/Jackson owner. Phil still plays Jacksons. The next, horrible photo is of Carmen Electra and Phil (excuse/explanation below). I was just thinking that if I touched Phil's t-shirt and Carmen touched Phil's t-shirt then there's a possibility that my DNA touched Carmen Electra (woohoo!). Last, but not least was the Incredible Hulk, Lou Ferrigno. He was on his feet signing autographs all day. I couldn't get a good photo of him, so I decided to incorporate a paparazzi element (the camera). The guy is still huge at 59. Can you believe that this year he might start qualifying for senior discounts?!



After taking the picture of Carmen my battery died.

Wednesday, January 26, 2011

••◊ Converting 5D mark II footage from 30p to 24p, part 1

My bodybuilding documentary, Stronger, was shot during the interim while Canon was still scratching their heads about why 24p should be available on the 5D mark II. The thing is that I still want it to have that film motion 24p look, but at 30 frames/second the footage looks like video. What's worse is that the Canon firmware I was using at the time only supported 30.0 frames/second, not even the broadcast/DVD standard of 29.97 frames/second. So...what to do?

Last spring I met Shane Hurlbut and he showed a small group of us a sizzle reel of his feature film converted from 30p to 24p via the Twixtor software. The conversion looked downright decent. My budget is obviously much, much smaller (i.e. non-existant). After a little research I found that Adobe After Effects CS5 has a feature called "Pixel Motion," which isn't exactly glaringly obvious at first, but does work to a certain degree. Pixel Motion analyzes the motion in the footage and generates in-between frames, thus doing frame rate conversion.

But first; a tangent...By default After Effects remaps from one frame rate to another by choosing the nearest source frame and duplicating it if necessary. We see this all the time with DVD's of film and 3:2 pulldown (in my case it would be 2:3 pullup?). This sort of processing can lead to choppy motion because you're either duplicating frames or throwing frames away. You won't see the difference in the still frames of the video, only when the video starts moving.

Another option is to use a process called frame blending, which overlaps two frames of video at one source frame rate (i.e. 30.0 frames/second) to create a new in-between frame timed to the desired frame rate (i.e. 23.976 frames/second). Premier Pro does this by default (After Effects can do it too) when you dump footage into the edit time line and the footage frame rate doesn't match the frame rate for the edit sequence. The frame blending algorithm basically overlays the nearest two source frames of your footage and sets the opacity of each source frame proportion to the time distance to the new frame of video. For instance if the 24p frame is 1/3rd of the way from source frame#1 and (obviously) 2/3rd's of the way to source frame#2, then the 24p frame is made up of 66.6% of source frame #1 and 33.3% of source frame #2. Does it work? ...yeah, kinda. The result of frame blending is shown in the first picture below. You can see that there are two frames overlaid and the result is a blurred in-between 24p frame. Any video with motion will blur in order to create the in between 24p frame. It doesn't look great when looking at stills, but the effect works for the most part when the video is in motion due to our persistance of vision. (Kinda looks like a 3D movie, huh?)



The second aforementioned "Pixel Motion" algorithm isn't available in Premier Pro, but is available in After Effects. Just to cut to the chase - the picture below shows the result of an After Effects render with Pixel Motion. Pixel Motion is similar to Twixtor and Apple's Motion software ("Optical Flow") algorithm. Does it work...again, yeah kinda. For the purposes of this post I'm just going to talk about how to enable "Pixel Motion. The idea here is that the software analyzes the two nearest frames of source footage and creates a morphed in-between frame that in theory represents what the picture would look like if it was shot at the edit sequence frame rate. Massive difference in quality, eh?



OK, so how does Pixel Motion work? First, create an After Effects project with the desired frame rate. In this case it's 23.976 frames/second (i.e. 24p). Then import your footage (File->Import) and drag it into the timeline. See the two red arrows in the picture below. The upper one enables frame blending when you click on the icon. The lower one enables either frame blending or Pixel Motion. When you click on the lower check box enough times such that it's a smooth slash from lower left to upper right then Pixel Motion is enabled. In fact, if you keep clicking on this box until you see a jaggie line from upper left to lower right you'll see the frame blending algorithm and you can easily compare the two algorithms. When the check box is blank that means frame duplication is in effect.



The final action is to render out your video in After Effects. I found that if I just imported the After Effects project into Premier that is not only takes a long, long, long time to render out the video but it also has a bug at edits where the wrong underlying track on the edit time line can appear for short bursts with Premier Pro CS5 (version 5.0.3 bug!). The safest thing was to just render out and take the degradation of second generation footage. I don't have enough storage to do raw TIFF sequences, so high quality H.264 did an OK job for me. I could then take the video and import it back into Premier to do my editing.

The major question is, does it look like film 24p motion? Well...again, kinda. Sometimes it looks film-ish, sometimes it looks a little video-ie. However, I'll put the emphasis on the "little" here. In my opinion it looks a lot more like 24p film motion now. One key thing that I didn't know to do at the time is to use a 1/50th shutter speed on the camera when shooting at 30p if I want to convert the footage. Motion blur is a component of the 24p film look, however most of this film was shot with 1/60th shutter speed. What I wish I knew back then...

So... is Pixel Motion perfect? It's very good, but not perfect by a long shot. Neither is Twixtor nor Optical Flow. I'm going to cover that subject with part 2 of this tutorial. I'll also cover how I got around a major problem with Pixel Motion, or at least sufficiently hid the flaw.

Sunday, January 23, 2011

••◊ "Stronger" Bodybuilding Documentary - The Director's Cut

I shot a documentary about a bodybuilder in October of 2009, then started editing as time allowed in December in hopes to make the deadline for the Seattle International Film Festival. Foolish as I was, I produced a product that I never really felt was complete. It was 23 minutes, no music or polish, but told a respectably compelling story of David Patterson, a bodybuilder who's been to the rodeo more than a few times.

So over the holiday break in December 2010 I decided to revisit the footage. A year of learning and perspective is a good thing in this case. Sometimes you can go back and do things differently. The SIFF deadline has come and gone, but that's OK. I finally produced the film I intended to.

So what changed? It's just over half the initial length, re-edited, music and color correction added, converted to 24p, and given a more compelling story. The dialog has been (mostly) fixed. What I wish I had back then is some of the simple film making equipment I have now, like a follow focus and field monitor, along with a steadicam and shoulder rig. I'm also more aware of audio quality now, but that's always a struggle when you're trying to be a one man band

Overall, it's a feeling of slight redemption. I present "Stronger - the director's cut."




Click on the image above to be taken to Vimeo to view the film in HD!

Monday, January 10, 2011

••◊ noise reduction on HDSLR footage

Technically, this post should be called noise "redirection" on HDSLR footage but we'll get to that in a bit. About a year and a half ago I shot a documentary with the Canon 5D mark II firmware in it's infancy. No 24p, no manual exposure, no manual audio, obviously not the best way to do it. On top of that I was in Washington State in September/October with cloudy light and no light kit. The answer...dah tah dah...ISO 3200/6400! My footage still has those rainbows of random CMOS imager noise ruining an otherwise nice image.

So you now know my dilemma. I want to use the footage but it looks like it was shot on a minio flip with a DoF adapter. After Effects has a noise reduction effect, but it tends to smooth over details and create "shiny plastic" looking video - nasty. There's also Neat Video, but that uses a similar low pass filter algorithm causing detail loss and it's currently $99 - I wanna be cheap here.

I've said it before here; film grain is supposedly "artistic" right? I mean you see it in just about every projected film. So what I decided to do was to turn that grotesquely colorful CMOS imager noise into film grain. How, you may ask? Our eyes see image details in luma (brightness), not chroma (color). Every image compression algorithm takes advantage of that and throws away lots of color information. So the first step to redirecting noise is to convert the image to a color space that separates luma information from chroma information, so I can keep the detail while operating on the CMOS characteristic colorful noise. After Effects can do a conversion from RGB to YCbCr using the "Channel Combiner" Effect (click on picture below). One thing I found out recently on an Adobe blog post is that when Adobe says 'YUV' they really mean YCbCr. For clarity I'm going to talk about the YUV conversion just so the overall topic doesn't get lost and the process is easier to follow in After Effects. The 'Y' channel represents brightness or luma. We don't want to touch that because that's our sharpness and detail. The 'U' and 'V' channels represent chroma, or color information. So for the "From" parameter select "RGB to YUV" in the drop down box. Now we are in a working space where we can operate on chroma separately from luma we can start chroma noise redirection to somthing that looks like film grain.



As with any noise reduction algorithm we're going to apply a filter to filter out the noise. I'm not an image processing expert, but I do know from raw image conversion that if I blur the chroma channels my remaining luma noise tends to look monochromatic (the holy grail of film grain). So the next effect I'm going to apply is "Channel Blur." Since we are already in "YUV", 'Y' equates to the "Red" Channel and 'U' and 'V' equate to the "Blue" and "Green" channels. We can now look at the noise in each of the chroma channels by using the little button with the three RGB circles below the preview window (see first picture in sequence below). I know it's next to impossible to see the noise with the small pictures in the main section of this blog post, so I recommend you click on the pictures to enlarge them. What you will see with the larger pictures is that the noise on the "Green" (i.e. 'U' or Cb) channel looks reasonably OK. The noise on the "Blue" (i.e. 'V' or Cr) channel looks terrible. Let's low pass filter that noise using the Channel Blur effect. This is done to taste, so to speak. I found a reasonable value for this footage is a "Blue Blurriness" of 30.



The final step is to convert YUV back to RGB. I just used that same Channel Combiner effect, but selected "YUV to RGB" as the "From" parameter. Nothing special otherwise. The image below shows the before and after. Again, you'll likely need to click on the picture to enlarge it enough to see the effect. The noise is still there in the "After" image, but it's been filtered and redirected to look like monochromatic film grain instead of colorful CMOS noise. Notice that unlike the "Reduce Noise" effect there is still reasonable detail in the fabric of his sweatshirt because we never touched the luma channel.



That mostly saved the footage. My half-ass first attempts at audio actually required more fixing. What's left is a conversion from 30.0 frames/sec to 23.976 frames/sec (i.e. 24p conversion). Turns out that After Effects can be quite adept at that as well, but more on the frame rate conversion topic later.

Monday, January 3, 2011

••◊ Continuing the topic of being cheap, meet CCMixter

When you're trying to do a video project, such as a documentary, it's often just a labor of love. Nobody is getting paid, and there's no commercial use. However, sites like Youtube regularly scan videos for copyright violations where someone has their cat dancing to Lady Gaga, then automatically replace the existing soundtrack your choice of cheese-ball keyboard demo song. Meet your new best friend, ccmixter.org.




At CCmixter.org you can find music that's available for non-commercial use and available under a Creative Commons license. As their logo says, "download, sample, cut up, share." Some well known artists, such as Chuck D (Public Enemy), Fort Minor, the Beastie Boys, and My Morning Jacket, have contributed work that can be remixed and reused on a non-commercial basis with credit given to the original artist. Essentially, it's for guys like me who want to work on their little projects without any commercial aspirations.

The music I used for the short Geminid video came from ccmixter.org. They have a fairly simplistic search system, of which I generally use the tag cloud. Sometimes you hit the correct song, sometimes you just can't find what you want. The selection is strongly in favor of those looking for keyboard-y beeps and bops, but some acoustic music is available.




Speaking of beeps and bops...there's also mobygratis.com. Unlike CCmixter, Moby created a simple site with just four pages of song-lets that can be used free for non-commercial use or licensed for independent film. Procedes go to the human society. The music selection here is generally slow paced and electronica oriented.

I won't even get into the topic of "fair use" as a documentary film maker because I haven't found anyone who understands it. Record companies believe that there's no such thing as fair use and documentary film makers often think anything captured live is fair use. What worse is that record companies are completely blind when it comes to licensing their copyrighted songs to millions upon millions of "small" projects (wedding videos, student films, documentary shorts, podcasts, how-to's...etc). They won't even return your calls. Moby is way ahead of the curve in this respect.

So check out ccmixter and mobygratis next time you need a song for a project. Results can be quite satisfying.