You are currently browsing the category archive for the ‘Movie Magic’ category.

A close friend of mine recently told me, in the politest way possible, that maybe it’s time to pronounce my most recent film project dead and move on. When I repeated this to my wife, she said, “I could kiss him for saying that. I’m not going to, but I could kiss him for saying that.” (sorry, man). Well, I think he’s right. Here’s what I learned from making this short 3 minute film. Or in other words, tips to producing a good short film.

The Good:

1) Storyboards created for the film helped ensure the story was understandable from a purely pictorial perspective.

2) The animatic was even more useful to estimating the pace of the story.

3) Verbally directing the acting. Speaking an actor through the emotions, as if shooting a silent film really helped get the right visuals.

4) The story works. It has flow, you can understand what the action is. That’s all good.

The Bad:

1) Create a plan view of the action for every sequence. This consists of a diagram a bit like a football play diagram. Then you can easily pick camera angles, and draw your storyboards.

2) Obtain shooting locations far in advance. Having to change all of your shots last minute due to being moved to a different location makes keeping shot continuity a real chore.

3) Be flexible during filming. Trying to cram the exact script down your actors’ throats makes for a very long day for them. Try leaving some fudge room and being accepting of different variations.

4) Have the storyboards and script on set and well organized. Being able to refer to them in an instant is really handy.

5) The actors didn’t really like acting. Well, one of them did, but the other (who I don’t think had acted before) had a tough time with the conditions on set.

6) Shoot the right pacing on set. Trying to get the timing right on set just has to be easier than attempting to fix it in post. I say after having tried to fix the timing a simple walk in post. This is where a field monitor would come in handy to review the shots immediately after filming them and check to make sure everything looked good. I may need to participate in a professional shoot to learn an effective way to do this.

The Ugly:

1) Plan out every detail of the special effects/visual effect shots. This includes a plan for obtaining the necessary equipment and rental estimates. It might even be handy to have a property master instead of doing all of the hauling myself, but that would cost more rental money. I shot the main actor “flying” without a leaf blower onset to add some “wind”. As a result, he really looks like he’s just hanging there instead of floating.

2) Use a good camera. The next camera I shoot with will need to have the ability to turn off the auto exposure feature. After wrangling with the exposure of otherwise perfectly fine shots, this has been beaten into my head.

3) Use an HD camera. The extra resolution gives me so much more latitude in recovering bad camera motion, bad focus, bad exposure, etc., if necessary, during post production.

4) Use a camera stabilizer. I’m currently planning on building a hard drive gyroscope stabilized steadicam. Should be fun!

5) Rehearse in costume. That will give a good indication of the final effect on film. Had I rehearsed the flying shot with the actor in costume, I’d have known that we needed to cut holes in the shirt to hide the rock climbing harness we used.

Advertisements

This is the establishing shot for my short film “Keep Your Feet on the Ground”:

Unfortunately, the consumer camcorder I used had no way to turn off the autoexposure feature of the camera, so as the camera pans down, the roof overexposes to white and the shadows under the eaves drop to black.  After a couple failed experiments, I was able to correct the shot.  Here’s the final version:

My first whack at this was just to try to color correct it.  But where the roof is exposed to white, you can’t color correct lost detail back into that area.  So since this shot was almost a nodal pan, with no action other than the camera movement, I decided I could create a panorama in Photoshop and then fake the camera move in Shake.  This is what I did.

1) I stitched together the best parts of several frames into on Panorama.  (Details on this step can be found in several tutorials across netland, so I’ll just illustrate the process with a simple, useless diagram.)

Some of the frames I used to assemble the panorama.

Some of the frames I used to assemble the panorama.

2) Next, I imported the panorama into Shake.  After fiddling with image brightness, I noticed some mistakes I made when stitching the images together, so I Quickpaint-ed them away.

3) I then added the camera movement using a Pan node.  This took some tweaking to get smooth camera movement and avoid looking off the edges of the panorama.

4) Now this isn’t a normal shot. The last frame of this pan is the first frame of a camera dolly in to the window. So the last frame of the shot couldn’t be changed from the original. I forgot that when I was making the panorama, and had painted all over the bottom part of the panorama. So I had to remake the panorama, being careful not to touch the bottom part this time.

5) The shot looked pretty good at this point, but I noticed the background looked like it should be warping a bit as the camera moved to better match the feel of the original shot. So I added a Lens Warp node. While it improved the shot, it also created several problems. It warped some of the panorama edges into view, and affected that last frame of the shot again.

6) I got around affecting the last frame by fading out the kappa value in the Lens Warp node over time:

Here you can see I faded the Lens Warp kappa value over time.

Here you can see I faded the Lens Warp kappa value over time.

7) To fix the edges coming into view caused by the Lens Warp node, I AGAIN had to remake the panorama in Photoshop, this time cloning in extra sky and bricks on the edges to make things work. (That’s the third time, for those of you not counting.)

The Completed Panorama

The Completed Panorama

8 ) Since the panorama isn’t video, the shot was missing video noise like a real video clip would have. I used a Film Grain node to analyze the sky on the original clip, replicate that noise pattern, and applied it to my panorama pan. Then I decided it didn’t look right, spent about an hour manually fiddling with the values, compared my manual film grain to the original clip, and decided the Film Grain analysis was way better than my poor attempts. Trust the Film Grain node analysis.

9) Finally, just to make SURE that the final frame matched up with the original footage, I actually spliced the panorama shot and the original clip together, making the original clip visible on the last frame. Ops OK. That means everything looked good, yo.

Here's the complete shake tree I used to create the shot.

Here's the complete shake tree I used to create the shot.

During my playful experiments with Fourier Transforms, I discovered that you can overlay an image on top of a Fourier transformed image, and freely convert back and forth from Fourier to regular space. Each transform would descramble the previously FFT’d image. So in short, you can hide the FFT of one image in another image.

Here’s the message I want to hide:

The Secret Message!

The Secret Message!

I’ve put it in the bottom half of the picture so it will still be intelligible after being transformed. Here’s the FFT of my secret message:

The Fourier Transform of My Secret Message

The Fourier Transform of My Secret Message

My first test overlaid the FFT over a color wheel, but it’s still a bit obvious:

The Secret Message overlaid on a test image.

The Secret Message overlaid on a test image.

If we scroll the FFT over to the corners and scale down the brightness, it’s much harder to spot when overlaid with the zucchini-in-a-bottle image:

The secret message offset to hide it better, overlaid on a normal photo.

The secret message offset to hide it better, overlaid on a normal photo.

You can barely spot it. Some information is lost when the image is saved in the .tif format, but after we scale the brightness back up, we get this:

The actual use decoded message (rescued from a zucchini-in-a-bottle)

The actual use decoded message (rescued from a zucchini-in-a-bottle)

Here’s my whole shake tree for both the test image and the final masterfully masterfool zucchini-hiding-a-secret image. Cheers!

Lossless Shake Message Tree

Lossless Shake Message Tree

Masterfool Zucchini Secret Message Shake Tree

Masterfool Zucchini Secret Message Shake Tree


(“Bravia” commercial clip Copyright SONY.)

Temporal Median Filters work like magic for removing small unwanted objects from video . . . like hairs and dust. I’ve often wished I had a Temporal Median Filter to remove unwanted noise from my videos. I always figured that Shake was flexible enough to do the job with just the available nodes, but I couldn’t figure out how to do the job until recently. I’ve made a macro out of my work which you can download here. Technical details follow.

I tried and failed with the LayerX node because it can’t do pixel-wise comparisons, and the TimeX node can only work with one FileIn node. But the ColorX node can do the job . . . if you can feed it the right information. Here’s what you do:

1) Create 3 FileIn nodes all with the same video source.
2) Edit the Timing>Timeshift parameters of one FileIn node -1, and another to 1.
3) Reorder the three video sources so that each only has the red channel. For video source 2, the red channel is reordered to the green channel, and for source 3, the red channel is reorder to the blue channel.
4) Combine all three video streams with a couple of IAdd nodes.
5) Add a ColorX node with this formula in the Red channel:
r+g+b-Max3(r,g,b)-Min3(r,g,b)

Red Temporal Median Filtering Shake Node Tree

Red Temporal Median Filtering Shake Node Tree

6) Repeat steps 3-5 for the green and blue channels.
7) Layer all the data together with some reorder and add nodes.

Complete Temporal Median Filtering Shake Node Tree

Complete Temporal Median Filtering Shake Node Tree

And if you don’t care to do all that, here’s a picture of the simple tree you’ll need to make after using my macro:

Temporal Median Filter Shake Node Tree

Temporal Median Filter Shake Node Tree

You can download the complete MedianTime macro here: MedianTime Filter at Creative Crash