As part of the AFF621: Remaking the Physical module in the MA for Digital Humanities (NUIM), students were assigned to do 3D recording exercises for the purpose of evaluating their skills in digital recording and to assess their abilities to carry out effective workflows for 3D recording cultural heritage projects. Students were requested to focus on two case studies for the digital recordings. The first case study involved the capturing of a prehistoric vessel within a museum setting, the second case study involved the capture of a statue in the outdoors. Students were requested to capture the objects using a digital camera, and process them using photogrammetry software and online tools to produce 3D models. Additionally, students were asked to submit a series of blog posts to describe their efforts, and to forward a final 3D model for 3D printing.
The choice of vessel used for the first case study was a bipartite bowl, from the Early Bronze Age, and was found in a cist grave in Liscooly, Co. Donegal in 2004. The vessel is documented by the National Museum of Ireland as file no. 1A/48/04 and registered as 2004E505:5. The images are courtesy of the National Museum of Ireland (Archaeology) who supported the project and provided access to the artefact. In approaching the first case study, I examined 3D recording and photogrammetry in the context of cultural heritage and discussed this in Post 1. I observed that the creation of a 3D object from 2D images is not merely about the production of a final 3D model, rather, it entails numerous and varied actions from planning the approach, designing a workflow, documenting the capturing and processing procedures as well as ensuring for the preservation of the data and the 3D model for the future.
In Post 2, I recounted the visit to the museum, and how I felt under-equipped due to a lack of preparation in researching the artefact that I proposed to digitally record. While I rectified this afterwards, there was a lesson learned and I will certainly be more prepared and spend some time in researching a subject for any future digital recordings in a museum or similar institution. I detailed the capturing process in Post 3, and outlined the faults with the captured data in terms of poor camera settings, and also some photographs which were not “Good”. Post 4 looked at the processing of the photographs to create a 3D model using Agisoft PhotoScan Professional, and detailed the various attempts that were made, due to issues with the captured data.
Thereafter, in Post 5, 123D Catch (Autodesk) was examined as a processing method to create 3D models. This post also demonstrated similar issues related to the captured data, but also in my approach. Additionally, I highlighted that the 123D Catch application seems to have an error in processing a capture scene from the manual stitching process. Finally, as mentioned in Post 2- Cahill and Sikora present information on the Liscooly bipartite bowl (2004E505:5) via 2D images and a written description, and while I was optimistic that I could create a 3D model to add to the record, it simply did not happen this time. Perhaps, in the future I will get another opportunity to do the capturing again, and if so, I would pay much more attention in attaining the best photographs possible.
After abandoning an attempt to do a capturing of a statue in the local graveyard, I then chose to capture a statue in my mother’s garden for the second case study. In Post 6, I discussed the type of camera used, the planned angles, issues with the sun, and how I attempted to attain good photographs. Thereafter, Post 7 described the processing phase with Agisoft PhotoScan Professional, which I deemed as a success due to the efforts made in the capturing stage and with the masking of the images. Also, in Post 7, I discussed 123D Catch for the processing of the photographs of the statue, and how it became non-responsive. In Post 8, I examined how 3D printing has advanced from being a costly and complicated process to being accessible and cost-effective in a few decades and how it is now used in the cultural heritage sector. I also created an STL file from the 3D model of the statue made in Agisoft PhotoScan for the purpose of 3D printing.
On reflection, this task was challenging, frustrating and extremely time consuming, but certainly worth the effort to learn about 3D recording through a hands-on approach. Thus, from this experience, I concur with Redmondino and El-Hakim that despite the advances of technology “the generation of a precise and photo-realistic computer model of a complex object still requires considerable effort” (Remondino and El- Hakim 269). Additionally, as was evident in this assignment, careful attention needs to be given to the capturing phase in order to produce “Good” photographs as this better enables the software algorithms to produce a more accurate 3D result.
Note: The 3D models produced for these case studies are available to view online in Autodesk 123D; Verold; and Sketchfab. The 123D Catch project folder for this assignment is available for download here.
- Remondino, Fabio, and Sabry El-Hakim. “Image-Based 3D Modelling: A Review.”Photogrammetric Record 21.115 (2006): 269-291. Academic Search Complete. Web. 16 Mar. 2015 <http://www.mdpi.com/2072-4292/3/6/1104/pdf>
- Cahill, Mary and Maeve Sikora. Breaking ground, finding graves— reports on the excavations of burials by the National Museum of Ireland, 1927–2006. Wordwell, 2012. Print.