 |
-photos become 3D |
| My photos to 3D | Help | Our services |
Frequently Asked Questions
About requirements of the photos.
General questions.
How does it work?
Answer: By photographing a surface or object from slightly different angles, there is in theory enough information to calculate a 3-dimensional model of the object (at least the parts visible on both photos), this is similar to what your brain does with the images from your two eyes. We use a very sophisticated computer algorithm to do this 3D calculation.
Why can't I see the back of my 3D-model?
Answer: Only surfaces visible on both photos will be part of the model, the computer algorithm used can not guess things that it can not see. The free service is limited to two photos. If you contact us, we can create 3D models from more photos (this is not free).
Why is there sometime garbage/noise in the generated 3D-model?
Answer: The computer algorithm sometimes gets confused about things in the pictures, this is especially true for large single colored surfaces. To minimize the garbage, be sure to follow the advice on how to take photos below.
What file formats are supported?
Answer: You will receive the 3D model in a zip file containing the following formats:
| Format | extension | |
| Anim8or | an8 | Used by the open source (free) program Anim8or. |
| Collada | dae | Used by google earth, use this format if you want to add the model to google earth. Can be viewed in the free program ColladaLoader. |
| VRML 2.0 | wrl | The "old" standard for 3D-objects on the web. Can be viewed in the free program Flux Player |
| X3D | x3d | The "new" standard for 3D-objects on the web. Can be viewed in the free program Flux Player |
Is 3DS supported?
Answer: Not directly, but the free program Anim8or can convert the an8 format to 3DS.
Can you import the 3D model into SketchUp?
Answer: This is not supported, but if you convert the model to 3DS first, then you can import the 3DS-file into SketchUp. (see the question above about 3DS)
How many "megapixels" do I need?
Answer: If your photos have 2 megapixels or better, you will most likely get good results if you have followed the instructions for how the photos should be taken. 1 megapixel photos will also work in most cases but it is not recommended to use photos with lower resolution.
Can I use a stereographic camera (a camera that takes two pictures simultaneously that simulate two eyes)?
Answer: Yes! A setereo camera is ideal for this application. Photos taken with a stereo camera will almost always yield good results
How can I improve the quality of my 3D model?
Answer: The quality of the generated 3D is completely dependent on that you have followed all the instructions. Read them again and see if you might have missed one. The most common mistake is to have a large angle between the photos. The angle should be very tiny (look at the examples if you are not sure how this could look).
How to take photos that can be converted to 3D models
Important: Photos can not be converted to 3D if these instructions are not followed. All of them has to be followed.
- The photos has to contain the same surfaces. (se figure 1 below)
- The object you photograph should not move. Move the camera instead.
Movements in the object between the photo shots will confuse the calculation algorithm and make the result look funny or not creating a result at all. If you are photographing people, ask them to stand still like statues in between shots. - Use a good light source.
This will minimize the risk of creating fuzzy and noisy photos. Sunlight is ideal, but if you use a lamp, do not move it in between photo shots. The light source has to be still between photo shots (just like the object you photograph). - Do not use a flash (use good a good light source instead).
- The two photos should ideally be taken within an angle of 5-15 degrees from each other (se figure 1 below). Note that this is a very tiny angle and the most common mistake is to use a much larger angle making the resulting 3D-model look bad.
- Make sure the object/surface you are interested in covers the majority of the photos.
This ensures that the object you are interested in is not missed. And you will get more polygons on the object of interest. - Avoid reflective surfaces.
Reflections in a reflective surface might confuse the 3D-generation. - Photos should have the same scale. Do not change zoom level in between photo shots.
- Only color photos will work (do not use black&white photos).
Se our examples to watch how the pair of photos could look. And when you are ready, use our form to start!
|
|
About the free versions .
Free version (free, poor quality, normal quality)
Is it really free?
Answer: Yes.
Why is it free?
Answer: We are doing this for free mainly as a sort of advertising to attract potential customers to our commercial business (see below).
What limitations does the free versions have compared to what you can do?
Answer:
| Poor quality version (free) | Normal quality (free) | What we can do (hire us as consultants) |
|
| Max Number of photos | 2 | 2 | (2-∞) unlimited |
| Max polygons | 1,800 | 20,000 | Unlimited, can be millions if that is needed. |
| Computation time | 50 to 100 seconds | 2 to 4 minutes | Depends on number of photos and details. Can be anything from an hour to a week. |
| Texture | Texture from one of the photos | Texture from one of the photos | Texture from all photos. (mixed textures or texture from the photo that fits best at every location.) |
| Feature extraction | Finds the biggest surface common to both photos (may produce bad results if no big surface is common to both photos). | Finds the 3 biggest surfaces common to both photos, often enough to produce good results. | Any number of details and features can be extracted. |
| Black&White photo support | No, only color photos supported. | No, only color photos supported. | Yes, Black&White photos require more complex computations that we can perform if it is needed. |
| Price | $0 (free) | $0 (free) | Depends on level of detail, number of photos etc. Contact us to discuss prices. |
| Comment: | Extremely important to follow the instructions for any useful result to be obtained. | Extremely important to follow the instructions for any useful result to be obtained. | If you want to take the photos yourself, we can assist you to get the best results. Or We can take the photos for you with our professional equipment for best results. |
Contact information
If you have any questions not answered here, please send an email to: info@photo-to-3d.com
Hire us as consultants
Send a short description of your project to info@photo-to-3d.com and we will write to you back with details of what we can do. More details on this page.
(c) 2007 Dogduck software, info@dogduck.com