Thursday, October 1, 2009

Trial two- Photo modeler

Trial two - After the first trial was so unsuccessful I decided to try to model only a portion of the wall. I went back to the building and took three more photo's focusing on the deep windows cut into the front left hand corner of the building. These are the images I used.





This time there was enough corresponding points to solve camera stations for all three photos. I marked and referenced as many points as I could but soon discovered I still did not have enough images to mark and reference all the faces of the window reveals. When I mapped textured surfaces onto the half complete 3d model it did not work properly. The finished result was an improvement but still unsuccessful... I will have to return again to take more photo's. This software is proving quite challenging to use and I am still not certain I will get a successful trial. I can not how ever dismiss the capabilities of the software so quickly, perhaps with more practice and understanding it will prove to work well....

You can see from the above image that I again underestimated the amount of corresponding points needed to create a completed 3d model of the structure. The 3d textures mapped onto the incomplete 3d model may possible be reflections from the windows in the photo's. I will have to go back and take several photo's to make sure I have at least two photo's showing every point.