This post is also available in: Español
What is LDRI?
Before you know what is a HDRi, I will say that is a LDRI. It’s nothing more than any image we can see on the internet or photos that we have done with a conventional camera and is called Low Dimanic Range image (LDRI or just LDR: low dynamic range) that contains minimal information of light and shadow it is necessary to represent our photo.
A LDRI file is used in websites, slideshows, TV, print, etc., it can be .jgp, .png, .tiff, .bmp, .gif, format, etc..
I will show you an example with the following image:
I have taken a picture and see how the object is closer clearly shows, however what is behind this too bright without coming to appreciate what is good.
I will try to fix this in the next picture:
Now with the same camera I made another picture trying to make the background look good, as in the previous photo was brightly lit, as shown, what is behind the window looks good but our object is closer this very dark.
That area of the first image that is too bright in photographic terms is said to Overexposed, the area of the second photo that is too dark is that it is Underexposed, photos of both the place that looks good is said to properly Exposed
These two photos above lack of light information to look good, the first picture is missing information shadows, the second photo lacks lighting information, so images are low-level information (low dynamic range).
But …. and if we take the two photos only part that looks good and put them together? that’s where HDRi comes in.
So what is Hdri?
As I showed above what is LDRI, the High Dynamic Range image (HDRi) is the opposite, that means it is an image that contai1ns information stored both lights and shadows which is able to capture with a camera photos.
This range is captured with a camera shown in “Steps” and is represented by the acronym EV.
A HDR or HDRi file is the same and usually carries a .hdr or .exr some may be also .raw, .tiff and others.
At present the HDR as such can not be displayed by any monitor, TV screen, or any device that projects images since even these lack the technology to display all the information of a hdr.
Knowing a little about photography
I’ll explain a bit more about photography oriented for HDR imaging. If you’ve ever had a professional or semi-professional camera in your hands you will have seen that there is a photo to the automatic option and the manual option. With the automatic option the camera chooses his values apparently lighting should have our picture and make photo, but this option is not worth to make a HDR. The best way is to take the HDR with the manual option, because with that you can control the EV steps ie can take a series of consecutive shots to capture lighting and shadows.
In photography there is the exposure time (shutter speed) which is nothing but having the control to overexpose, expose correctly or underexpose an image.
If you look behind the viewfinder of a professional camera or semi-professional camera you’ll see something like this:
Among many things, seen in the bottom left a number which in this case is 200 (1/200), this value indicates the exposure time will take your photo, this value you can manually increase or decrease to overexpose or underexpose as you see fit.
Below you’ll see a chart that normally have exposure cameras:
This table can vary from one camera to another, but in general each value is an exposure time and jump from one value to another contiguous counts as “step” or 1 EV (1/60 to 1/125 is a step or 1 EV and 2 to 4 is a step or 1 EV, etc etc) so we have 1/15 to 8 are 8 EV’s. We can say that there are intermediate values between one step and another, as shown in the photo above the 200 would be a middle of the 125 and 250. Therefore to create a HDRi you need to do a sequence of several shots with different exposure times and then join them into one and get a high dynamic range.
In the tutorial parameters and camera settings I explain more about shutter speed, aperture, distance, focal length, ISO, depth of field, and many tips.
How many Ev’s need to create a Hdri?
If you like this post and want to keep reading the rest of the content, please click on one of the buttons of social networks (facebook, twitter or google+) to show hidden content:
Thank you 🙂
Now that you know that you must capture a series of photos to create an HDR I will show an example. I have this image sequence:
6 photos with different exposure times ranging from 1/1000 to 1/15 so loose these 6 pictures do not serve me must unite into one and for this there are many free and paid programs like Photomatrix, HDRShop, Hugin, Photoshop, etc.
While data of lights and shadows you have a HDR can get more out of it when loaded in a 3D application. However the amount of EV’s a HDR depend on the own environment that we capture with the camera, look at these examples:
The following picture I have obtained a 17 HDRi EV’s.
This other picture I’ve obtained a HDRi 14 EV’s.
What happens with HDRi with more EV than they should have?
As I have shown that a LDRI lacked information of light and shadow as a HDR with more EV’s than you need what you’re doing is loading unnecessary information. Also if a HDR has few EV’s of those who should not have come to be a LDRI but would lack complete information.
Hence a HDR should have the amount of EV’s necessary since these give good lighting, shadows and good reflexes.
As I will show below, when less EV, lighting is lost and therefore sharp shadows is lost and reflects mainly the light source is lost.
Tone Mapping, hdr advantage
The images and panoramas shown in this and all the web tutorials are LDRI since a HDRi not be displayed by any monitor, TV screen, much less for a printer. If you put in any internet search engine images the word “HDR” or “HDRi” the result will be like the following images:
These images HDRi many call should actually be called Tone mapping because HDRi as mentioned can not be displayed as such.
Tone mapping is a process that is done after obtaining the HDRi in which you can properly expose what is overexposed and/or underexposed, saturate, dramatizing the contrasts, etc., then burn it with LDR extension therefore loses all EV’s.
Unlike a HDR to light a 3D scene in which the capture of many EV’s need, to create a Tonemapping only need at least 3 photos of different exposures, eg 3 shots -2, 0, +2. or sometimes with a single shot in RAW.
Recognizing a good Hdr
On the internet there are many sites that offer HDR’s (. Hdr or. Exr to CGi) free and paid, but many are not of good quality, even acceptable quality. A HDRi must have these four qualities:
When a hdri carries a light source as powerful as the sun or even a light bulb must generate defined shadows, when it generates a good catch because of EV’s
The following example obtained free HDRi “Large Balcony” generates these shades:
In a Hdri panorama elements that comprise it must clearly be seen, both the objects that are closer to the camera as they are further away, otherwise it will affect if you want to use as a reflection or background image in our 3D application .
To use an HDR reflecting or background (backplate) the picture must have good resolution in pixel, depending on the size of the HDR render to be used as a reflection should be no less than 3000 x 1500 pixels and background image even more.
A good capture of the EV’s will make our HDR lighting projecting good without straining many parameters that increase the render time or adding extra lighting.
Unlike HDR for reflections or background image, the picture to illuminate a scene goes well with HDR’s 800 x 400 pixels, HDR’s with more than 1500 x 750 pixels to illuminate create unnecessary burdens to the application memory 3d.
When the photographic capture of EV’s is correct, give very marked, especially in those materials where the reflectance is not extreme reflective glare. Being very highlighted the brightness in the area where the light source is almost negligible and the rest of the panorama.
Even in bright or light colors.
Resemblance to reality:
With a good HDR panorama captured a likeness to reality is accurate. Look at this comparison of a real photo with a render in which we used the HDRi “The orange home with Natural light”.
Transforming a LDR to HDR?
The LDR carry information of light and shadow that occupy 8 bits per channel (8 bits R = Red, 8 bits G = Green, 8bit B = blue) ie each pixel of the image has the option of representing 16,777,216 colors while a Hdri containing information includes 32 bits (4,294,967,296 colors each pixel). Before and I showed the LDRI have extensions jpg, png, etc, etc, and is also known that an image editor can change its length by any other (from jpg to png, tif to jpg, jpg to hdr, etc.) So you can create an HDR of a LDR? in practice yes you can eg open in an image editor an LDR panorama 8-bit extension .jpg and convert it to 32bit and then save it with a .hdr but really you get that? a fake HDR, there are many of these online, with the extension .hdr or. exr but with internal information of a LDR. Let me give an example that occurred to me:
As you see in this graphic representation, 8 green cubes fill all fields orange panel, but to make the conversion from orange to blue panel has 32 cells, many of them are empty, it is information that is not covered by green cubes .
Another example more according to pictures, the “Histogram”. In any image editor you can see the histogram of an image, which is a graphical representation of how much light or shade has our image, notes an example:
With that notes how it has changed the histogram by transforming a LDR in HDR shown above on the right side of the histogram many spaces that have not been able to complete information
HDRi is the joining in a single file in a consecutive series of photographs taken at different exposure times to capture the widest information of light and shadow.
I hope you serve these concepts to answer questions about the dynamic range.
Have you had any experience creating a HDRi or Tone mapping? Do you know anything else you want to bring to this tutorial?
Please leave us your comments.
SUBSCRIBE TO BLOG
[smlsubform prepend=”Enter your email to receive notification of new tutorials or downloads” showname=1 nametxt=”” nameholder=”Name (Optional)” emailtxt=”” emailholder=”email” showsubmit=true submittxt=”SUBSCRIBE” jsthanks=true thankyou=”Thank you very much for subscribing to the blog, as soon as I have new news I will send it to you in the mail. Greetings and Good day!”]