I spent some time this past weekend mounting a couple of prints from my trip to my ancestral Montana home last summer for my office at work. As an avid amateur photographer, I’m always trying new techniques to produce the strongest photographs possible. One of my favorite techniques that I use quite often is known as HDR (high dynamic range) photography in pro and serious amateur photography circles. It is a relatively new technique, though not a new idea, made possible by the strong digital processing options available to digital photographers today.
Here’s the thing. If we are to consider the action of the pupil in the eye adjusting to varying light, the eye has almost 24 f-stops of dynamic range–said another way, we can see a starlit sky in the dead of the arctic night all the way to sun blazing off of a placid lake on the equator at noon-an increase of one f-stop essentially doubling the brightness perceived. But even at a static pupil aperture, the eye can still detect between 11 and 14 f-stops of dynamic range, and we use the dynamic range of that wonderful instrument daily to see detail that no past or present camera (including film) technology can capture.
The state of the art dynamic range available from the sensors on high end digital cameras or video cameras today is about 8.5 f-stops–material outside of that shows up as either straight black or straight white. So while an eye can look at a scene and see 11-14 f-stops of dynamic range, a digital camera/video camera will only detect slightly more than 8 f-stops, and because this is a logarithmic scale, this is a significant difference in performance.
The technique I presently use in my digital photography is to bracket three quick exposures, one under exposed, one properly exposed, and the final over exposed. Both my cameras can be set to do this automatically, and both support “rapid fire” exposures–up to 8 frames per second. So this can be done quickly as one is shooting the picture(s), but (and here’s the rub) these three negatives must then be combined digitally using computer processing techniques to create a single high dynamic range image. In other words, almost 64 megabits of data from multiple negatives must be processed to create a single 20 megapixel image in HDR. The results are stunning–detail that looks realistic to the eye when properly done. Like anything else, this technique can be taken to the extreme and not all HDR pictures are particularly life-like.
So what does this have to do with my TV? Most current video displays have the raw ability (if not the actual circuitry) to display dynamic range that simply isn’t available in most video content. It is possible that the “next big thing” in video will be getting the content, storage, transmission streams, etc. to move to true high fidelity video (not just high definition) by increasing the dynamic range to something closer to what the human eye can perceive. My sense is that this element may be more important to video cognoscenti than 3D in terms of actually enhancing the video experience.
Next week, I’ll discuss the fine points of this future technology and the ramifications to the industry should it be adopted. In the meantime, take a look at this sample of HDR video–understand it’s highly compressed and just a rough approximation of what is possible, but very intriguing nonetheless.
Until next week….




