From 3D digital models
Nowadays it is getting very cheap and very easy to obtain 3D digital models (by which I am referring to 3D triangular meshes) of specimens for morphometric analysis. For example, using micro-CT, the structure of interest (e.g. skull) can be virtually dissected and saved as a 3D mesh file (e.g. ply file). Or you can use a surface scanner to capture a digital model of just the outer surface of a specimen. It is even possible to build digital models using photos from your phone (e.g. using 123D Catch).
Once you have the 3D model files, you can use a few different software applications to collect the 3D landmark data. Your choice of software will depend on the types of landmarks you wish to digitize and the complexity of the models you have.
IDAV Landmark Editor | This is the standard free software for digitizing landmarks on 3D models.It is Microsoft Windows Only (but works OK via emulator on a mac). With this software you can place landmarks on the surface, as well as inside complex structures. You can also use the curve function to define a curve in 3D space along which equidistantly spaced semilandmarks can be placed. The most attractive feature of Landmark Editor is the semi-automatic digitizing; you place all the landmarks on a single mesh (called the atlas), then you place the first 4 landmarks on the next mesh, correspond the two and a cloud of landmarks are automatically positioned on the second mesh, which the user shuffles into position. This removes the common digitizing error of mis-ordering landmarks.
Volumetric data comes from a variety of sources. For example, you may be using micro-CT or MRI for visualising your objects of interest. Or you may have digitized serial section slices. Or you may have slice data from grinding down fossils. From all of these stacks of images, you can make threshold the slices to make 3D meshes as above. But you may not want to make an isosurface from a single density structure in the scan (e.g. bone). Therefore there are alternatives to working with the volumetric data saved as images (slices):
Checkpoint by Stratovan | This software is commercial and only available for Microsoft Windows. It comes from the original authors of IDAV Landmark Editor mentioned above. It works very similarly to Landmark Editor, accept that it allows the input of image files that are as stacks of CT/MRI slices (DICOM, TIFF, JPEG and others).
The perhaps most traditional way to collect 3D landmark data is using a microscribe. The best feature of this method is that the microscribe is portable and can be taken to museums or to the field for data collection. The digitizing arm is connected to a computer and is operated by a pedal - when the pedal is pushed, the 3D coordinates of the position of the digitizing arm tip are placed in a spreadsheet. One limitation is that there is a small size limit to what can be digitized, and mistakes cannot be rectified later, since once the object is moved, the coordinate system cannot be recreated.
Photogrammetry is the science of making measurements from photographs. There are a few ways to take 3D landmark data from 2D representations. Below I highlight a new and very exciting package by Aaron Olsen.
StereoMorph package | This package in the R Statistical Environment is a way of using photographs taken from different angles to collect 3D landmark data. The blog for this package describes very clearly how the method works. The most attractive aspect of this method is that it is very cheap to do and quick to use.