Right now there are an increasing number of tools to use photos to aid the 3D design process. ReCap360 and 123D Catch fom Autodesk are two popular options, along with PhotoModeler and the late ImageModeler from Realviz. Most modern matchmoving software can also create point clouds from photos.
Adobe’s newest tool fits into this already highly populated 2D to 3D world, but it also stands out due to its unique design. Instead of using a series of photos to generate the 3D information, Menglei Chai, a PhD student from Zhejiang University, and a team of Adobe Research scientists¹ have worked out a way to intelligently derive a 3D head from a single photo. While the tool is only able to reconstruct the visible areas of the face, essentially creating a digital relief, the results are quite impressive.
Currently, Adobe has released only a sneak peek of the technology at their annual MAX conference and will be presenting a tech paper at SIGGRAPH Asia. Whether or not it will be included in Photoshop and when remains to be seen. For now, take a look at the examples and express your thoughts below.