Will intelligent Cameras augment 3D Cinematography?
The Future of 3D Cinematography:
When contemplating the future of stereoscopic 3D movies and 3D at home, it is sometimes good to fantasize (I do sometimes live in a fantasy world, driven by technology) about where Digital Cinematography is heading. There is more evidence today than ever before of how we have come to rely on technology to make mainstream decisions for us, even at the cost of human artistic intuition.
Today you can run an open wager that any Cinematographer from a modern film school, will be quite out of place if asked to shoot with traditional 35mm film and without “video assist” as was done in the old days. Of course Digital Technology *is* evolution, and this article is all for it and about it. With that in mind let’s see how technology and modern Cinematography could evolve.
Some forward looking thoughts to consider are, what will such an effect be on established “movie industry” roles? Will technology usurp the older established profiles of AC, (assistant cameraperson) 2nd AC, or indeed of Cinematographer? These assumptions or predictions are only to provoke thought and are not without base, and while there is no replacement for Human talent, we all know that technology has either replaced or augmented Humans in many mainstream areas…it is now moving into the world of Movie Making.
Technology Assisted 3D Camera Framing:
In the first image in this article, professional and budding photographers and cinematographers will instantly notice the ‘rule of thirds‘ when framing a picture or a scene in a movie. There is no hard and fast rule to be enforced about the rule of thirds and it is broken many a time with artistic intent. However it’s a rule of thumb and if programmed into a camera could always turn up perfect framing without much second thought for the movie maker.
The Director or indie camera operator / director could then decide whether to override it or not. How could this be implemented and in what scenarios?
In a narrative movie, it could be used for perfect framing of people. This would work with technology such as Face Detection; the core algorithms that give features such as smile-shutter detection in consumer grade point and shoot cameras.
For 3D framing, appropriate tweaking of such an algorithm could lead to nicely framed 3D shots, iterations of which could be used keeping the inter-axial (distance between left and right lens) calculation of 3D cameras in mind, and thus displaying an online grid also taking into account the off-set that will occur between left and right cameras as interaxial distance changes between or during a shot.
How about an intelligent algorithm that tracks a “Two shot” or uses the automated focus motors of a 3D Camera to frame the perfect OTS (over the shoulder) shot for a long mise-en-scene type shot? without stereo window violations?
Depth of Field isolation – or- Background De-Focus for 3D:
Still borrowing from technology used in consumer cameras and without going into Pro features (coming up below), another powerful feature can be borrowed from new Sony Technology: Background De-focus.
This is a feature that 3D Movie directors would die for! The ability to completely de-focus the background and have the subject in a Circle of Isolation, so that the audiences will concentrate on the main Protagonist or characters in the scene, instead of getting distracted and straying into the rich visuals in 3D of the background. We all know that rack-focus or bad use of depth of field, can make audiences suffer immense eye-strain or headaches over the course of a movie, if the depth of field leads to semi de-focussed backgrounds. The brain would instruct the eyes to unsuccessfully fuse such imagery, which can lead to eye strain and equally important – detract from the Directors story telling.
The Sony background-defocus technology if it *could* be incorporated in a 3D camera would intelligently blur out the background, making for sharp and clearly demarcated 3D focus in a scene.
Shooting RAW; Augmenting the DoPs skill-set?
Back in the days of 35 mm film, there were no second chances at getting the “look” of a scene right. A cinematographer had to know exactly what ISO film-stock to use, lighting, what filters to pop into the Matte Box and much more. There were chemical post processing after-effects that could be employed to changing the “look” to a certain extent, but those were limited as to what can be done with Digital processing today.
Shooting RAW or capturing the digital data straight off the sensor in the camera, allows for almost completely engineering the “look” later in post-production, from White Balance, to Color to exposure. Recovering shadow detail and Highlights is not un-heard off. Something that could not be done previously and needed a completely experienced DoP and Lighting Crew.
The whole concept of D.I. or Digital Intermediate is just that…shooting on film and then scanning it into computers to color correct and post process. With Digital Camera sensors now equal in size and pixel count to 35 mm film, coupled with the fact that these Digital Cameras now cost less than a few thousand dollars, the whole pros and cons of Digital versus Film is changing.
Modern Digital DSLR cameras with extremely high ISO and no visible “digital noise” (versus film grain) can capture images that film ISO cannot. This means a whole different game changing scenario for Cinematography. While it is probably preposterous to suggest that a $2000 DSLR matches to a high end film camera or even high end Digital Cinema camera.
Arguments are made that even though DSLR sensor sizes are as big as Vista Vision, the actual compression bitrate is too low, and the processing actually drops every alternative line of visual information, resulting in moire interference patterns and other artifacts… the question has to be asked: How long before these get addressed? If the main stream manufacturers protect their high-end products, how long before a rebel company is born and follow the same success formula as RED?
What does Digital Cinematography Technology mean to the future of Movie Making?:
I think it’s safe to say that every single filmmaker and photographer has always dreamed of cameras that can see what our naked eyes can see. This time these cameras can actually see more. Sure – they may not have the dynamic ranges of our eyes just yet – but they see more than my naked eyes can see in low light.
Period.
And that’s qualifies as a paradigm shift in my book.
– Vincent LaForet (who shot the video above on a Canon 5D Mk II Dslr. with no additional lighting)
The video above was shot on a Canon video DSLR. A camera that costs a fraction of what a typical Digital Cinema Camera costs. While that is worth considering, what is worth contemplating is the more far reaching implications that the video reveals:
- It was shot ONLY with available light. To put this into perspective, imagine a movie being shot with no Lighting Crew or lighting gear being ported around. (very forward thinking statement…but plausible )
- The ISO on that camera even at high setting completely impresses the established gurus of Cinematography in Hollywood, as the Zacuto Film versus Dslr test showed.
- The camera can already output uncompressed RAW image data from the sensor over HDMI, and all that is needed is a firmware update to take advantage of this
- How long before a Video DSLR comes out with a genlock option, making it the most lightweight 3D rig available?
Analog Mindset in a Digital Generation:
One important thing to remember is TALENT cannot be replaced with Artificial Intelligence or Technology. That will (at least for now) be the domain of a real person. If we look at the music industry, it’s no wonder that talent from the 80′s and the 90′s are still producing hit music. Meanwhile music making technology is available to every teenager or bedroom musician.
While it may be a comforting thought, we should also realize that Evolution or (erosion), whatever we want to call it, has progressively degenerated and rendered this generations ears in-sensitive to the difference between 192 kbps sampled sound and the nuances that can be heard and felt when listening to CD a couple decades ago…and Vinyl and tape before that.
This should start Movie Professionals thinking. With Cameras getting more intelligent will there be a need for Focus pullers? Lighting crew? bulky equipment and more?
Advances in Digital Technology when combined with Artifical Intelligence, will undoubtedly create intelligent Cameras that will take out the technical guess-work in production, and put storytelling tools in the hands of anyone who cares to experiment.
R&D teams in the electronics industry are already researching new technology for upcoming 3D capable cameras. Are we ready to adapt… and evolve?




















