1

I have a video and would like to create what is called a Photo Montage like this: (combining several photos of the same skier like below)

enter image description here

I have searched on the Internet and some definition of "Montage" is different from what's shown here. I know some Android phone has this feature when taking a video. But right now I already have a video and like to create this layered time lapse effect on either Mac OS X or iPhone/iOS.

4
  • Keynote's instant alpha would be the easiest way to isolate the skier without using Photoshop or advanced video processing software. You'd have a lot of manual work to paste the object you have isolated on the correct path.
    – bmike
    Commented Dec 23, 2013 at 18:46
  • if I do paste them, then the position will be subject to errors for how I place them at which x and y coordinates. so it will be best if the placement is done by the computer Commented Dec 23, 2013 at 19:50
  • This seems like a lot of hassle, but you definitely could have proper placement that way: stack all the images together, use Instant Alpha, and mask them. Alternatively, Photoshop works on Mac, and you can import video frames as layers, which you could mask yourself. I expect that's what the creator of your example here used.
    – Tuesday
    Commented Dec 24, 2013 at 17:59
  • There should be a better software solution. If done manually, it might be able to be done in 20 minutes. And then if you need to do another photo, then another 20 minutes. Versus if the software can handle it, it should be able to be done within a few seconds. Commented Dec 27, 2013 at 11:18

1 Answer 1

1

Cool idea! Here's how I'd do it using Final Cut Pro X (or Compressor 4). This won't help if you're looking for something automated. Combination is already been discussed in the comments, so I'll offer more detail on generating stills prior to combining them.

1. How to extract still images from the video using Final Cut Pro X

Here's how to get several stills from a video using Final Cut Pro X. My answer assumes you've imported the video file and added it to a project.

First, add a destination.

Final Cut Pro X doesn't automatically offer you the option of exporting all the frames in a video project. You must "activate" this feature. Open Preferences and select the Destinations panel. Click "Add destination" and choose "Image sequence": Here's what Apple has to say about FCP X destinations.

Second, export or "share" the movie.

Click the "File" menu, select "Share", and click "Export Image Sequence". In the modal export dialogue box, you can adjust file format under the "Settings" tab. I recommend defaulting to TIFF (a lossless format that won't add any compression artifacts).

2. Select the images you want to combine.

Judging from the example above, you're looking for images that are spaced evenly. Perhaps frames 10, 15, 20, 25, 30 will be the ones you need. Pay attention to the quality of each individual shot. If frame 15 is blurry, then try 14 or 16. This could be done using Preview, the Finder, iPhoto, Aperture, or your layer-friendly photo editing software of choice.

3. Combine the images.

As I acknowledged above, I don't have anything to add to the comments already made. Use your layer-competent photo-editing software of choice. Depending on how much the camera moved while filming, you may need to stretch, blend, and rotate each layer.

Considerations

Consideration 1: Interlaced video yields messy stills.

It has been a long time since I've worked with interlaced video, so I don't have much to say. Just know it's an issue with some video cameras, specifically when the resolution is "i" not "p" (e.g., 720i). For example, check out the image on the right:

progressive vs. interlaced

Consideration 2: Video frames have wayyy lower resolution than still photos.

The bottom line here is that frames extracted from a video will not give you nearly as much detail to work with as still photographs. The resulting "montage"/time lapse will be much lower resolution. (I recognize that the content in question has already been filmed, but for posterity I wanted to point this out.) Unless you're shooting with an incredibly high resolution video camera, like a 4k camera, the number of pixels in one frame of video is much smaller than the number of pixels in a still photograph. 1080p video, the highest resolution on DSLRs, is 1920 x 1080, or 1.98 megapixels. Contrast that to 18 megapixel stills on the same DSLR (e.g. Canon 60D).

Consideration 3: The more the video camera moves, the more difficult it will be to combine your shots.

So use a tripod.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .