A have many question on your article. Please, could you answer?

Methods:

1. Dummies normally do not have texture, why did you use them? In my experience it is not a good method to test optimal configurations with photogrammetry.

2. What is the criteria to evaluate as acceptable reconstruction those obtained with only 18 cameras? Fig. 1A illustrates scans that reconstruct a human skeleton, that is an hollow body having very thin bones. Although I am using photogrammetry for face and body scanning since many years with high quality APS-C sensors and calibrated optics mounted on digital reflex cameras, using same software adopted by the authors, it seems to me quite impossible to obtain the results reported in figure 1A using single cameras (not coupled in a stereo scheme) positioned so far from the subject and only on the top, at 2.3 m of height. Please, could you explain better these remarkable results?

3. Please, dealing with photogrammetry it is necessary to specify optical characteristics of the camera (sensor image area, focal length, type of lens, focal ratio, shutting time). Could you provide these data?

4. Normally in the body scanners the cameras are positioned at different heights on the same column (see suggested references). Why you adopted a different scheme? A circular frame having 4.8 m of diameter is a different frame respect to the one reported in figure 1B. Please could you explain?

Data acquisition 5. The sample has only 6 persons (4 male and 2 female); it seems too limited respect to the scope of the works. Which was the type on inclusion criteria adopted?

1. Why the authors used clothes, being the purpose of the paper the calculation of BSP? It could be better to use directly the texture of the skin of subjects wearing only slips or bikini, avoiding over estimations.

2. It is difficult to believe that the quality of the result reported in figure 2 was obtained using only 18 PI cameras with fixed focus positioned at 2.3 m. PI camera has a very small sensor, 3.76 x 2.74 mm, and the photos show that the subject occupies approximately 1/5 of the sensor area (1 Mpixel). The low quality of the sensor and the very limited number of pixel involved lead to a very low quality 3D scanning reconstruction, due to the triangulation algorithms. Could you share the original photos?

3. Agisoft Photoscan can scale the project directly using scales or markers positioned on the subjects, or doing the external camera positions calibration and using the function "Ground Control" (now “Reference” in version 1.1). Why you adopted a different procedure?

4. The method adopted for scaling is not clear: please, could you explain it?

Results 10. Why shoes were used, and not only sock (for better foot surface estimation)?

1. I do not understand the sense of the comparison of measures of scans done on 4 males and 2 females, acquired dressed, with the average values of the literature. You do not believe that the sample is not adequate?

2. The body reconstruction of figure 8 seems to me that was not obtained using the described scanner. The quality is too high, but mainly there are details on undercuts that are not visible from cameras positioned only on the top, and there are so fine details on the hand and foot fingers. The undercut of the chin also can be obtained only with a "sub mento vertex" point of view. Could you explain or share the original photos?

3. I have serious doubles that 18 PI camera are sufficient. Many references reports different experiences, also using the same type of cameras: a. Richard Garsthagen www.pi3ddscan.com b. Cam Peterson, Pann Ajjimaporn, Jiaoni Wang, Jeremy Straub, and Scott Kerlin. "A Raspberry Pi-Based 3D Scanner" University of North Dakota Graduate School Scholarly Forum (2014). c. Available at: http://works.bepress.com/jeremy_straub/174
   d. Jeremy Straub * and Scott Kerlin, Development of a Large, Low-Cost, Instant 3D Scanner - Technologies 2014, 2(2), 76-95; doi:10.3390/technologies2020076 - See more at: http://www.mdpi.com/2227-7080/2/2/76/htm#sthash.gNgm28rS.dpuf http://www.mdpi.com/2227-7080/2/2/76/htm e. James Hobson, An affordable full body studio grade 3D scanner, http://hackaday.com/2014/03/07/an-affordable-full-body-studio-grade-3d-scanner/

4. The resolution adopted seems to me too low for still images, but extremely insufficient for motion (video images).