Can’t seem to get something to synth right? Curious about what lens is best? Ask your fellow Photosynthers here.
I started using photosynth to map out and provide a walk through of my house and yard. Would it be possible to create a Synth from a video without going to images first.
And as a second question, would it be possible to download the "points" (or a subset of them) as a 3D file ?
Regarding video synthing, the Photosynth team has admitted that they have experimented with it, even before releasing Photosynth in August of 2008, but nothing has been released yet.
For now, it is pretty simple to use something like VirtualDub or Quicktime Pro (depending on the format of your video) to convert your video to an image sequence in one simple step. Putting every image or a subset (such as every other or every tenth frame) into Photosynth can be easily accomplished by viewing the folder as thumbnails, allowing the file selector to be wide enough to fit the appropriate number of photos in each row and then select one of those columns to put in the synther. (e.g. two columns for every other photo; ten columns if you wanted to select every tenth frame)
Downloading the pointcloud is fairly simple nowadays due to some Photosynth users' hard work. Download their tool here: http://pspcexporter.codeplex.com/
Many different 3D Model tools can open the output formats supported by that exporter, but my favorite so far is Meshlab. You can download it here:
You may notice that in synths where not all the photos synthed together (especially where there are different clusters of photos visible when viewing your synth in 2D mode) that when you use the pointcloud exporter, there are also multiple pointclouds a single synth. Usually you are just interested in the biggest one (which is usually the first) but they aren't always listed from largest to smallest, so if there are many, be sure to scroll down to make sure you aren't skipping any of the larger secondary pointclouds.