1,438 karma · joined May 1, 2020
For Unreal I used a few methods but mostly conventional photogrammetry incorporating the lidar.
Ultimately, I'm hoping that downloading the file or some type of pixel streaming for web until the nerfs or splats -- or whatever follows them -- works out.
I was wondering about this too: they've found high levels of mercury in the water supply at Maya cities and believe now it contributed to the eventual collapse: https://arstechnica.com/science/2020/06/mercury-and-algal-bl...
Or here's the trailer for the project in Unreal: https://www.youtube.com/watch?v=hlNgpG9X7mc
I have a lot of work keeping up with games, it's true--games are expensive to build and aiming at photorealism art style continually looks dated quickly while stylized graphics, less so. I'm trying to fundraise to build this game currently, but it's a tough sell. Educational games don't do well on Steam, so right now, I'm just distributing through my website as I build. The small income this provides helps me contribute back to the modern Egyptian Egyptologists that are excavating and documenting their own culture though.
The latest Game Science title, Black Myth Wukong, does an awesome job with 3d captures of Chinese monuments and bringing the mythology and history to life.
That's an impressive huge capture!
So in the virtual tour, you're seeing 360 imagery from the cameras and a lower resolution version of the 3d capture data, optimized for web. The lower res mesh from the scanner is transparent in first-person view mode so users get cursor effects on top of the 360 image.
For film, PBS sent out a documentary crew, and they wanted me to render some footage of the full tunnel system, so I exported the e57 pointcloud data from Matterport and rendered the clips they needed in Unreal. It should be coming out soon with "In the Americas."
For me, the Maya have always been important because they're our history and our stories in the Americas (I'm from the US) -- more than the greco-roman mythology I grew up with. They grew corn and love ball games. Their stories are more directly our stories, and their struggles are our struggles.
Not to be political, but they also kind of wiped themselves out by large scale environmental collapse, and the jungle is filled with their undiscovered monuments. There's still so much to learn.
People really geek out on how much the Maya knew about astronomy too -- they shot archaeoastronomy docs twice while I was working on site. Richard Feynman even helped decipher Maya glyphs and writes about it in "Surely you're joking Mr Feynman". He gave a lecture also if the audio file can be checked out somehow: https://collections.archives.caltech.edu/repositories/2/acce...
I did take some scans after the collapse! After we'd dug ourselves out and crawled out on our bellies, I went back with Polycam. The collapsed section we dug through was comparatively small, maybe 4-5 meters: Section 1: https://poly.cam/capture/4BB863F2-1CC3-46E3-8BDB-232EE3057BD...? (you can see where we crawled out to the intersection here -- the whole intersection had ceiling collapse, but only the section we dug out through was fully covered). Section 2: https://poly.cam/capture/3C5BB7BD-5FC9-4C00-AE1C-84E0544C51C...
We're just lucky it wasn't a rocky ceiling that fell, that would've been much worse.
The team taking care of the tunnels is doing an amazing job with the resources they have, and they're continually backfilling tunnels now and maintaining the ones that are there. It took us about an hour to dig out.
You can compare to the intersection in the matterport version in the same vicinity: https://my.matterport.com/show/?m=r5BR6K6Qxix&ss=338&sr=-.21...
I don't want to editorialize too much, but at that moment we were totally brothers--I was still early with Spanish, and the language, country, age differences fell away, and we dug ourselves out.
For terrestrial lidar, you can avoid the laser and cameras by following a circular path opposite the device sensors or hide behind a rock or tunnel corner, etc.
I used an iPhone 15 and Insta360 one rs with the Leica lenses -- without tripod. I actually went through and edited the humans (including myself) out of every shot, but you can see where I missed some. I'm hoping to go back through and fix them. I put up other tours on the website just from 360 video, and they're much less edited, so I'm still hiding at the bottom of most of the shots until I can edit myself out.
I aligned the 360 images with Metashape, exported the cameras.xml file with camera transforms/rotations, and converted these to my own json format with a simple 360 virtual tour viewer that I created w/Three.js.
For navigating between tour points, to try to approximate a similar motion affect as Matterport and Google Streetview, I somewhat unsuccessfully project the output of a cubecamera as a rendertexture on the dollhouse lidar of the tomb. I'm not happy with its current state but had to keep moving forward, and I appreciate much more the engineering that went into Matterport's Showcase frontend especially.
You can manipulate the splats in viewer, kind of similar to a pointcloud, something to the effect of https://twitter.com/parazar/status/1713618668788109357 but there are a bunch around social media.
This has an entrypoint to the research and various projects so far: https://github.com/MrNeRF/awesome-3D-gaussian-splatting
The Time snake in the depths in the Netherworld still mystifies me.. and the Barque of the Earth! Wonders. OsirisNet has good interpretations for going deeper: https://osirisnet.net/tombes/pharaons/ramses1/e_ramses1_01.h...
And the Darnell's Netherworld Books book is awesome: https://www.amazon.com/Ancient-Egyptian-Netherworld-Books-Wr...
Or for audio, Dominic Perry's amazing podcast with the 2 part narration of the Book of Gates: https://www.egyptianhistorypodcast.com/175-in-the-hall-of-th...
AR in these tombs would be amazing!
I also added a light bulb button to the top too to toggle the lights and guided mode.
Threejs has been great to work with. It's pretty simple orbitcontrols but the occluders on the dollhouse for the cursor in first-person view are a little more complex. I hope to share the codebase, but it's definitely still a sketch instead of the full picture--right now with a lot of adhoc parts for this specific tour for the vfx.