2 posts were split to a new topic: Forgotten Worlds - Global Heritage Inventory
Here I submit you the link to an archaeological site on which we are working, that can provide you an example for a possible temporary solution to the problem of scale representation of a photogrammetric model (scan) and also to the geographical orientation of it.
It is represented in Orto vision / Camera Field Of View 0 °.
See what you think of it.
A very interesting and necessary topic of debate.
It works OK with hotspots but free navigation doesn't work nicely because of the long distance. First person view won't work. Some basic measurement tools would be handy or scale bars can be added for references e.g. https://sketchfab.com/institute_of_archaeology_uw
Uomdigitisation is right in the fact that the model doesn't work properly in first person mode, in orbit mode we find no problems with navigation although.
Look closely at the notes 7 and 8 in the model, where the scale is displayed, knowing that each box measures 10 x 10m, the whole board becomes a frame for metric reference.
x, y, z
Grouping models in folders is nice... would be handy to be able to reorder the folder as well, as the order in which the models are presented can be part of the context of the grouping.
Agreed. It would be very useful to have the ability to check on a scale of objects without having to interpret dimensions. The way I did it for one of my model is by adding a xyz try-cylinder of 1m norm. But it is not really pretty and hinders the view in some directions as it is not user-manageable.
I am not sure how the creation of a point occurs in 3d especially if you allow for clickable points outside the model (at what does the mouse point when you click but an infinite line?) but I'm sure the S-team will think of something.
I don't understand the need for a length bigger than what is provided. There is already the possibility of inserting an url to link to a descriptor. Maybe allowing to add a description file (uploading a file) on the design page?
That post was a year old - we have since increased the maximum length of the model description.
a short contribution regarding the map of 3D models which was discussed here a year ago. One GIS student did his practice in our company, so we decided to prepare a simple web application linked a map directly to the cultural heritage models on Sketchfab. The result is here www.3d-mon.com
It is a simple pilot for viewing 3D models and their location, without any extended styles, but still quite interesting, I think. We have got lot of ideas, how to improve this pilot:
Would it be benefitial or helpful for any of you and should we continue with this project?
awesome @3djohny - really nice project, and I spy some of my scans in there too
those all sound like great plans - maybe hit up @corentin about his experience / work on culturemap.org (unforts this URL seems defunct now...)
what university are you / the student based at?
Very cool project. Mapping 3D scans is a great idea. I started Culturalmap with this ambition more than a year ago but did not have the time/ressources to make it a great website so I decided to close the site. I realized that something important is to give the ability to add metadata such as year of creation, origines, materials... It's useful data for scientists and allow them to better filter content.
The screenshots on the Culturalmap twitter page can give you an idea on how it looked like: https://twitter.com/culturalmaporg
Hey Bart - relatively new here... but really keen to push the envelope with what Sketchfab can already do!
I'm going to be experimenting with using a model on a projector this coming Tuesday. The annotations system works beautifully for presenting to a group already, being easy to negotiate a virtual 'tour' at the click of a button with various views all possible. What I've tried to do with one of my models (Atherton Hall - Digital Reconstruction) is to add additional shortened URL links to content I can click to while using this to present. You'll be able to see two examples in the model here:
What would be absolutely amazing is the ability to animate a orbiting camera path around a model which you could click on while you present, which would give the audience an additional wow factor (as if seeing a building that hasn't existed for 200 years isn't enough).
Joking aside, I'd be very keen to see this develop further. The possiblities of combining Sketchfab with Google Cardboard is something I'm incredibly excited about.
Are there any possibilities of linking Sketchfab models with Google Maps in the future? This could be incredible...
Mapping is going to be an obvious application for a lot of CH material. @bartv Rather than geotagging models in external systems perhaps it's time to add some extra metadata fields in Sketchfab for geotagging and provide an XML feed that can be harvested? I'm thinking of other possibilities like the 360Cities/Gigapan/Rumsey Historical Maps layers in Google Earth.
I'm following this discussion with great interest, lost of wonderful suggestions that I would find very useful as well!
I have some things to add to this that I believe could be beneficial for Sketchfab + Cultural Heritage Projects (but I'm sure also useful for other users).
Text annotations expansions: I'd like to have the option of having expandable / collapsible text annotations. I could see it being very useful to provide some basic information when first clicking on an annotation point (name, date, location, material, brief description for example) and then a button "show more" beneath that which can be clicked so the textbox expands to reveal supplementary information. This would provide more targeted annotations in my opinion.
Embedding / IFrames: When viewing a model in presentation mode an argument could be made that its best not to have to direct users outside this presentation for more information / material that could be embedded. By expanding and improving on the current possibilities with the Markdown frames for annotations & simple images a lot could be achieved in making Sketchfab a great way to present 3-D work with all the necessary information all in one place. Cl3ver does this in a brilliant way at the moment, sadly their pricing structure will not be suitable for many cultural heritage applications. see an example of one of their presentations here: http://3d.cl3ver.com/hDQC0?tryitlocation=2
Video: Inserting youtube / vimeo or similar video's within an IFrame like structure would be wonderful. This way additional information & animation and more can be presented without having to exit the model.
Images: Options to embed images are in my opinion one of the most important features that could really benefit from dedicated improvement in Sketchfab at the moment. Flexible sizes & transparencies would be great (comparing a site/model with a historic depiction of the same in an overlay for example).
Model in Model: In addition to embedding other formats I think there is a powerful argument that could be made to use what Sketchfab is so brilliant at (showing 3-D models) within a model as well. For example a model of an archaeological site could be shown on the first level, but then several of the artefacts found there could be embedded / tagged in such a way that an IFrame pops up allowing users to see that specific object as part of its context.
Audio: The ability to add audio (annotations) would be great when exploring models that deserve a lot of explanation. This could work as audio commentary (audio-tour) or even historical audio fragments, but I'm sure also nice as more atmospheric feature (playing some appropriate choral music when exploring a model of a church for instance). This would also overcome the limitations to the amount of characters that currently exist.
Floor-plan: An optional site-plan / floor-plan as a semi transparent overlay in a corner (preferably collapsible) could be very useful to show where in the model one is exactly, a bonus for any virtual tour. Especially when viewing buildings, interiors, Ruins this could help users navigate around, and it has the possibility to show the compass direction as well (which has been suggested by some other posters here). An example of how this could work can be seen here: http://createholo.com/how-to-add-a-floor-plan-to-your-holo/
Silhouette for scale: Some suggestions have been made to add a scaled ruler to the models so the scale can be visualised. I would like to add to that an option to include a standard sized human silhouette which I've found the general public relate to a lot more when they need to understand the size / volume of a 3-D model, especially if its a large site or of an architectural nature.
Hope these suggestions are useful.
Martin is a student of GIS at the Palacký University in Olomouc (CZ). We are a small company engaged mainly in 3D scanning and printing last 2 years, but I managed development of GIS before. So, when Martin needed GIS practice in some company, we decided to take him and try to join GIS functionality with 3D models.
I saw the discussion about Culturalmap, but unfortunately the project link was dead. But from twitter it seems like very similar project ...
Be alive, any geo-application needs the geotaggs done by an author (in this case the author of the 3D model) and they should be saved as attributes of the 3D model on the Sketchfab. Then 3D models can be used in any geo-apps like 3d-mon.com, Culturalmap or others.
To get the coordinates, it is the best for authors to have some simple map content directly on the Sketchfab - e.g. simple Google API window. They could click to the map on appropriate place and get the coordinates that should be stored with the 3D model.
It is not difficult to implement this functionality and it is very beneficial, from my point of view.
Geotagging in metadata fields and XML feed - YES! I would like to have it. The mentioned applications use external systems (Google) but implemented through API - e.g. Rumsey Historical Maps called it Georeferencer ... but it is based on Google API.
Yes, geotagging would be awesome. The challenge that we have is that each user segment has different requirements - CH people like geo tagging, architects like scale, etc. So we want to find a flexible solution that does not clutter our interface with long lists of options.
It's definitely on our radar, and I'll add another +1 for @3djohny
Can certainly appreciate the dilemna. GIS people probably want bounds, not a point.
Possibly list additional metadata in an accordion to save space and make it easier to select the required set. Should be possible to include the entire IPTC schema then
I don't think that bounds will be preferable features, maybe in the case of bigger scanned areas. But for small objects, statues, buildings ... points should be enough. Another possibility is to use semantic definition of the locality (adresses, names of cities, geographical names, ect.), but from my point of view the coordinates are preferable. And better just coordinates sooner then later the entire IPTC schema