Multi-Shots photography

Multi-shots technique permits to take pictures better then your camera best performances! Most people won’t even reach the full potential of their camera by using either low quality jpeg, not full resolution or by relying on the camera internal processing to do some adjustment, but using theses techniques you can acheive great results!

So let’s start by making one thing clear. It is impossible that your camera achieve the same computation performance than your computer. I run on battery and has to perform almost at real time.

Things like down sampling (not cropping), HDR, panorama are CPU intensive and can take from a few second to some minutes to my quad core computer. When my camera does it live, it is taking shortcut. Don’t the camera version when possible.

Now back to our main subject: the multi-shot techniques. They are they art of taking multiple picture to enhance the final results. Here is a list of techniques:

1-Panorama

It is probably the most known feature because it is found is all smart phone and permit to zoom out. Most Smartphone will down sample the picture in order to do it faster. As opposition to a real wide angle lens, the sensor might move during the process causing some distortion between the object close and the object far away. It works best, when all object are relatively far away.

Not only it permits to zoom out more, but it as also another hidden benefits when performed in post processing. It is to increase the resolution. If you use your zoom-in and then zoom-out, the result remains almost the same (except for depth of field). However you have increased your resolution!

2-HDR

HDR stands for High Dynamic Range. It consists of taking multiple times the same picture, with different exposure and permits to increase the bit depth. Bit depth is also a synonym of dark blacks and brighter whites. It is particularly useful for taking picture of highly contrasted subjects. However, as most monitors doesn’t render HDR information wet, it has to be scaled down to the device dynamic range. This process is called tone mapping. It is essentially decreasing the contrast, but most version will permit local contrast that sometime give bold edge effects that limits its effectiveness for realistic pictures.

3-Focus stacking

This one is not as common as the other one. It consists as taking the same picture with at various focus distance. Then each zone are evaluated for sharpness and are blended together to increase the depth. It very handy for macro photography but can be useful to increase the depth field of large sensor cameras. I have developed a process to automate this process with Lumix Wifi capable cameras. You can view it here.

It is also possible to do the opposite. That would be exaggerate the reduce depth of field of large sensor cameras.

4-Super Resolution

This process exploits small movement of the camera between multiple shot with the same subject and the same settings to create a higher resolution image. It is more frequently used in video where the image resolution is relative low. With today’s camera reaching resolution close to the lens/shutter precision limit it is of limited use.

5-Subject removal

While taking multiple shot of a scene it is possible to remove moving subjects. It can be via manual masking or averaging all pictures.

6-Noise Reduction

Again doing a simple average of the same picture took multiple times, this permit to cancel out the noise. This as limited use at low ISO, but for higher ISO it can be useful. For example, if you want to capture a moving subject at low light. If you can take multiple capture of the scene, you can cancel out the noise of the fixed object in the scene. For the moving subject, noise will probably be better then motion blur and can always be enhanced manually in post processing.

7-Long exposition

Most digital cameras have limited exposition time. Somethime it is not possible to reduce the amount of light entering in the sensor enough to achieve the desired results. To capture exposition longer then its limits it is possible to simulate that effect in processing using a “lighter” overlay between the pictures. It has the advantage to be more robust against camera movement (with picture alignment), but of course it is not much efficient to amplify weak signals.

8-Multi-Spectral

Using pure color filter and multiple monochrome pictures it is possible to take a picture that contains more than 3 primary colors. The result is more for scientific applications because it is not well supported in typical photo editors, but it can be great for exploring the world around us, creating mask or analysis.

I don't have a wide choice of pure color filter, but you can see some of my experement here...

9-Multi-Polarisation angle

Have a linear polarization filter? Try to capture various the same images at various polarization angles. This can be useful to select area of the picture based on its polarization angle.