Just to get this blog up and running, I am starting out with an easy post: An impressive time-lapse video of Las Vegas and a bit of background information.
First the film, 24 Hours of Neon, by Philip Bloom, uploaded on May 7th on Vimeo:
Philip Bloom is a professional filmmaker and an authority on filming with DSLR cameras, a technique that has exploded in usage since the introduction of the Canon EOS 5D MkII camera in 2008 – the first DSLR that featured full HD video recording. The large image sensors on DSLR cameras allow for a more professional-looking picture than what is possible with regular consumer video recorders, in particular a very narrow depth of field. For an example of that, have a look at one of his other films, Dublin’s People.
In this particular film, though, he did not make use of DSLR video. It is made up entirely of still photos shot on four different cameras during 5 days in a hotel room on The Strip in Las Vegas. There are two main techniques involved in the photographing; time-lapse and HDR (high dynamic range). I will just briefly explain them here, planning to write more detailed posts later about my own (more modest) experiences with these techniques.
In time-lapse, a long series of photos taken from the same location at fixed intervals are combined into a video, usually with somewhere between 10 and 30 photos (frames) per second. This gives a huge speed-up and a sense of time passing in no time. Everybody will be familiar with the clips of a blooming flower which are typical time-lapses. Other kinds of shots often seen are rolling clouds and street traffic. These require shorter intervals than the blooming flower, so alternatively they can be done by regular video in fast-forward, but using still photographs enables special effects such as the zooming or rotation seen in the Las Vegas film.
HDR tries to alleviate the poor dynamic range of digital image sensors relative to the human eye. Our eyes can simultaneously resolve a large range of light intensities, but image sensors, photographic film and computer monitors are much more limited. Therefore a single photograph cannot include details from both dark shadows and intense highlights. By taking several photos of the same scene with different exposure levels, shadow, midtone and highlight details can be captured separately and later combined in software to produce an image with a high dynamic range. In order to display that image on a standard monitor, the image must be converted to a low dynamic range (LDR) image file in a process called tone-mapping. The tone-mapping involves a lot of technical and artistic choices that can significantly alter the final look of the image – often the results look absurdly unrealistic; at other times they just look slightly surreal, as I would say is the case in this video (the HDR starts at about 1:30 min).
Combining the time-lapse and HDR techniques is an endeavor I could perhaps imagine trying for a single sequence, but never consider making a whole movie out of – at least until I had a massive empty hard drive and a week with nothing else to do. The amount of work that Philip has put into this project is almost as stunning as the resulting film.
Apart from being a great filmmaker, he also puts a lot of effort into sharing his techniques and teaching his craft, so his website is a great resource for inspiration and learning. I highly recommend you to go to his blog post, where he writes in detail about the making of the film. There is even a 36 minutes long commentary to go along with the video, as well as a Behind the Scenes video and a demonstration of the HDR/tone-mapping process.
I must say that it can be somewhat disheartening to see the massive amount of hardware and software that he has access to, but most of what he did in this film can be done on cheaper cameras and free software. Please hang on here for my upcoming posts on some of the methods I use for making time-lapse videos.