Creating Timelapse Videos – Tutorial

I’m likely to slowly migrate most of what’s on my main website to the WordPress format and figured I’d start with this tutorial rather than putting it over there.

I’ve been doing a fair bit of timelapse shooting of late and have had some questions about the process so thought I’d put together a short tutorial. This isn’t going to go into extensive depth on video editing as that’s really beyond the scope of what I’m wanting to outline here.

Timelapse is the opposite of high speed photography. In high speed photography, you capture images at a very fast frame rate and when played back at a normal frame rate, the action appears slowed down. In timelapse, you’re effectively lapsing time or skipping time. You capture at a slower frame rate and when played back at a normal rate, action appears sped up. If you’ve seen, for example, a video of the progression of a flower bud coming into bloom, that’s timelapse.

Let’s get the concept of a ‘normal’ frame rate taken care of first. Many of today’s digital video cameras capture video at 30 fps or 60 fps. Some also capture at 24 fps which is, historically, the rate at which cinema films are captured. When I talk about a ‘normal’ frame rate, I’m talking about 24 fps (also seen as 23.976 fps). This is the rate I use when creating timelapse video clips.

What equipment do you need for timelapse? Well, a camera is a good idea. Pretty much any camera that will allow for full manual or semi-manual (e.g., Av, Tv) operation will work. Fully automated cameras can be used but as with other types of photography, you lose some creative control of the final output. What else? An interval timer is also a useful tool. In an absolute technical sense you could manually release the shutter (with a cable release) and manually count between frames but that gets tedious. An interval timer (intervalometer) allows you to set up the parameters of the shoot in advance, set the timer to start then simply wait till the sequence is complete – no fuss, no muss.  Some cameras have good interval timers built in.  For those that don’t, you’re going to need something external.  Most camera makers offer a cable release/interval timer combination so check available accessories for your particular camera.  There are some third party options as well.  Phottix is a supplier of a variety of third party accessories, including interval timer/cable releases at reasonable prices.  If you’re a Canon or Nikon shooter and want a really slick tool, you can check out the Promote Control from Promote Systems.  It’s not inexpensive but the flexibility it offers is terrific.  For cameras that have minimal bracketing capability, the Promote Control allows you to override the in camera AEB.  It even allows you to combine AEB and interval timing for doing HDR timelapse work.  It’s a very cool piece of kit.  Another important thing for timelapse is memory space. You’re going to be shooting a lot of frames. You want to have enough memory to be able to capture the required number of shots for the clip you want to create. A tripod (or some other form of solid camera support) is also vital. You really can’t handhold for this kind of work, unless you’re doing some sort of moving sequence (i.e., shooting through a subway window or in a moving car). Of course you want enough battery power to keep the camera running for the number of shots and length of time you’ll be shooting.

What format should you shoot it? It’s really up to you. The larger the image size, the more card space is going to be taken up and the more card space you’ll need. Some will suggest shooting anything but RAW is foolish. While I’d agree for general photography, for timelapse it’s probably not as crucial. More important than format, I think, is image size. Full HD is 1920×1080. Shooting at larger file sizes – even JPEG – allows you to add some pan and/or zoom effects after the fact in editing and still be able to fill the screen. This can be desirable particularly if you’re a fan of the ‘Ken Burns Effect‘.  The other reason I’m less fussed about shooting RAW for timelapse work is that due to the nature of the intent – creating a video – when each individual frame is on screen for a split second and the dynamic movement of the video and the small size of the images, the benefits of RAW to still photography aren’t as crucial to timelapse video work.  You do have to be more careful with getting the white balance right and ensuring you take a more measured approach to editing if you shoot JPEG but shooting JPEG for timelapse you can create very high quality results.  If you’ve got the card space and like the enhanced flexibility then by all means shoot RAW.

What should be the spacing between shots? This will depend on a few things.  It’ll depend on how fluid you want the end video to be.  It’ll depend on how fast the action you’re shooting is.  It’ll depend on how long you want your video clip to be.  It’ll depend on how much card space you have.  It’ll depend on how much flexibility your interval timer gives you.  In general for more fluid motion in the video and if there’s a fair bit of action in the scene you’re shooting, a faster frame rate (narrower spacing) between shots is advisable.  If the scene is fairly static and/or you’re not as concerned about fluid motion, a slower frame rate (longer spacing) can be used.  Typically I’ll use anywhere from 1 fps to about 1 frame every 12 seconds.  More often than not, I’ll shoot at 1 frame every 3 or 6 seconds.

How many frames do I need to shoot? That’ll depend on the frame rate you shoot at, the playback frame rate (e.g., 24 fps) and how long you want the clip to be.  The simple math is clip length x playback rate = number of images.  So if I wanted a clip of 10 seconds playing back at 24 fps, I need 240 images.  The shoot duration math is # images/shots per minute = duration.  If I capture at a rate of 1 frame every 6 seconds that’s 10 frames/minute so I need to shoot for 24 minutes.  That assumes you import the images at the same frame rate you’re going to use as your playback rate.  For simplicity sake, that’s the basis we’ll work on.  If you want to get creative and use different import and playback rates, the math will change.  It also assumes you don’t do any stretching or compression of the clips in your video editing software.

OK, so you’ve gone out to shoot, now what do you do? Once you’ve got the images from the shoot onto your computer, you do whatever normal editing you’d do for any other image.  I use Lightroom for the bulk of my editing so making a change to one then syncing that change across all the others in the sequence is easy.  I’m not a Bridge user but I believe something similar can be done in Bridge.  If you’re working with another image editor, you’ll have to check the documentation for it specifically to see if you can sync edit changes across a series of images.  It certainly makes the work simpler and quicker.  The one place where you’ll need to work on a frame by frame basis is if you have to heal/clone any dust spots or other unwanted items.  This is particularly true in something like a sky where there can be light/cloud movement.  Doing a clone/heal on the first image and carrying it across all the others isn’t always going to work too well.  Keeping your lenses, any filters used and your camera sensor scrupulously clean is key to being able to avoid this time consuming and tedious chore (ask me how I know how tedious and time consuming it is 🙁 ).  Once you’ve made all your edits the last step is to create a set of images in numerical seqence.  If you’ve shot RAW, you’ll need to export them to JPEG.  Using Lightroom makes this task very simple.  In Bridge, I think the Tools>Batch Rename utility will do this for you.  If your images are still in numbered sequence as they came off the memory card and you’ve shot JPEG, you can use the camera numbering as your sequence.  If you’ve deleted any images in the sequence, you’ll either have to import multiple sequences or do a batch export/rename to create a new sequence.

Photoshop, since CS2 or Cs3 has included some video editing functionality.  To import your image sequence in PS, you go to File>Open, select the first file in the sequence, make sure Image Sequence is checked at the bottom and click Open.  See the screen capture below.  Note:  The video editing features are available in the Extended version of Photoshop, not the standard version.

Open Image Sequence in Photoshop (click for larger version)

Once you click Open, you’ll be presented with a dialogue box to select your frame rate.  Choose 23.976, 24 or 30 (whatever you’ve determined will be your default frame rate and what you used to calculate your shooting duration and number of frames required) from the dropdown menu and click OK.

The image sequence will then open and it’ll look like any other image  you open in PS.  The indication that it’s a video sequence will be on the image thumbnail which will show a small set of film frames in the lower right corner as in the screen capture below.

Image Sequence Indicator (click for larger version)

In the screen shot below, you’ll see I’ve turned on the Animation timeline via Window>Animation.  This allows you to see the timeline of your clip in the bottom of the screen.

With the sequence open in PS, you can make global edits via layers or directly on the video sequence just as you would with any other image.  These global edits will be applied to the entire video clip.  There are some more advanced editing capabilities for video in PS but those are beyond the scope of this introductory tutorial.  I may do future tutorials to cover some of those functionalities.  The last step in editing is to crop to your desired aspect ratio.  For HD that will be either 1920×1080 or 1280×720.  For SD it could be 640×480 for normal screen or 720×480 for widescreen.  If you lose too much real estate because your file size is too large, you can rez down the sequence the same as you would for a still image using Image>Image Size.  To save this step at the end of the editing process, you can batch crop/resize to the desired size in the batch export/rename step noted above.  Either one works.

I’ve done my edits, what next? Now you need to render out your video.  To do this, go to File>Export>Render Video.  You’ll be presented with a dialogue box that looks like the one below.

Video Render Setup Screen (click for larger version)

At the top you’ll select the destination for saving your video file.  In the File Options section, select MPEG-4 from the Quick Time Export dropdown menu.  MP4 is a good general purpose file type.  MP4 renders quickly and produces high quality clips at reasonably small file sizes compared to, for example, AVI.  It’s a preferred file type for video sharings sites like Vimeo and YouTube.  If you’re looking to make a DVD of your video, that’s a different issue and not within the scope of this introductory tutorial.  Click on the Settings button to bring up the screen below where you’ll configure your video.

Render Settings (click for larger version)

Choose the settings as they’re done in the screen capture above.  Lastly click on Video Options to bring up the screen below.

Video Options

In this screen, change Faster Encode to Best Quality.  This will slow down the render process somewhat but the resulting video will be of higher quality.  Click OK to get back to the main render screen.  Click Render and wait for the clip to complete.  Once that’s done you’re ready to upload it to a video sharing site like Vimeo or YouTube.  If you want to view your video on your computer, you may need to download a video player that’s capable of decoding mp4 video.  The free VLC Media Player will allow you to view your mp4 videos.

That’s it.  You’ve shot, edited and complied your timelapse video.  There are numerous other editing packages out there for video such as Sony Vegas, iMovie, Final Cut, Pinnacle, Adobe Premier and others.  Cost of these packages varies from about $100 for a very basic, bare bones option to well into 4 figures for a high end, very robust, professional video editor.  You can also do basic compiling and global correction with the free (Windows only) Virtualdub software.  The biggest downside of Virtualdub is that you can only save out files in AVI format.

I’ll do another part on timelapse video to encompass a cool sub-genre of timelapse known as HDR timelapse where HDR stills are used as the input files.  On the video page of my main site, you can see a sample timelapse clip that also includes HDR in the last 5 seconds.

Comments

comments

Powered by Facebook Comments

2 thoughts on “Creating Timelapse Videos – Tutorial

    • Sure, CHDK is a workaround that some people use. Some people don’t want to hack their firmware though and some don’t want to do it while the warranty is still in effect. There is a CHDK hack for some Canon DSLRs as well but I’m not sure if it includes this capability.

Comments are closed.