Week Summary 5/13

This week was very busy with tests/AP's and a slew of work to do on this project. I am happy with the progress for the Art Gallery Experience, and I hope that I can get it running on the Web by next week when I present. The Unity engine project will hopefully also be fixed by that time, because I want to ensure that Zoe and whomever else wants to work on this VR project next year has the resources and set-up ready to do so. It would also be very cool to publish the VR experience on the Oculus store for other people to be able to see all of the scans and environments that we have made. Next week I am gong to wrap things up and finish everything as best I can and present.

0 comments

"Meshes for Web" 5/12

I primarily focused on making the Achri-mesh assisted Gallery reconstruction today. The walls and simple geometry was easy enough to get down, but now I need to spend some time actually getting the Madge Tennent paintings in a usable format, including the borders and frames. Archi-mesh has proved to be a very useful tool.


0 comments

"Gallery Mesh and Drones" 5/8

Today I worked on creating some new Gallery mesh geometry, until Ilan and the rest of the drone guys helped me get some exterior shots of the elab for Reality Capture, that might soon come in handy. The pictures were a little out of focus when I dropped them in for reconstruction, so we might have to re-take some of the angles to get a good model.

0 comments

Week Summary of 5/6

This week I focused on recovering quite a bit of files from the accident. Furthermore, I got to talk to industry experts about VR/AR which was very helpful for the development process and to create a better virtual experience. I worked on cleaning/re-constructing the gallery mesh so that I can deploy the experience to WebGl, which allows for the VR experience to be represented as a standard 3d environment within the browser. Next week I plan on focusing to create a fully web deployed Gallery tour, and also fix the Unity accident, and prepare for presentations.

0 comments

"Gallery Experience" 5/5

Today I worked on creating a cleaner geometrical model for the Gallery mesh by taking measurements inside of Reality Capture and re-creating the walls of the Gallery in Blender. This allows for much, much cleaner geometry that will actually work well in WebGL so that the experience is viewable on the web. I used a plug-in for Blender called “Archi-Mesh” to quickly re-construct the basic geometry in a very simple and straightforward way.

0 comments

"Industry Connections" 5/3

Yesterday the Digital Journalism class that I'm in had a guest speaker, which is the lead of new idea development along with AR/VR experiences. This was a good opportunity to gain some insight on how to create a better story and experience in a VR world. The biggest take away was the tip on narration: never assume to user has completed a prior experience in the virtual world, have it so that no context is needed with transitions from one part/environment to the next. So, today I spent time brainstorming on how to make the Gallery tour deploy-able on the web so that we can have a showcase for the connection that was made.

0 comments

"Recovery and Tour" 5/1

Today I worked on recovering some of the 3D files that got lost with Unity. Good news is that we have all of the assets that were imported into Unity intact, so that means redoing all of the work is much, much easier. Zoe and I also had a chat with some foreign ambassadors along with the rest of the class about project work in a classroom and what students benefit from.


0 comments

"New Improvements and Mistakes" Week of 4/29

This week was both exciting and a little harsh. The new touch controllers are now working in Untiy after some efforts Friday evening. So there is progress on that front. However, due to the setback experienced Friday, by essentially deleting the entire Unity project, we are really not in a good position with presentations coming. Essentially what happened is somehow the Unity files were overridden with new ones, and so the progress that we had in the VR demo is gone. The assets, all of the scans, models, and textures, we still have. This means we just have to put the Unity experience back together, which is not such a big deal.
Next week's priority will be to get the VR experience back to how we can show something for.
It was hard on Zoe, but I think I managed to keep things sane and positive. I belive that this was a good learning experience, especially now that Zoe will be in charge of things next year with whomever is going to work on this in ISR for next year.
The touch controllers now work within a blank Unity project. This is actually a solution to the biggest problem that the now gone Vr demo had. So in a way, there is actually a diamond in the rough. Once we get the entire VR experience back together, it will also have a major improvement with how people interact inside of the VR space, which is very exciting and very good.

0 comments

"Setback and Learning Experience" 4/28

At the beginning of the day during ISR, we started with a conversation on project presentations and the general accomplishment of everyone in the elab. It was a very good conversation, and I think that I got a lot out of it. It is nice to see your work come together to this point.

However, sever minutes after that we found out that the Unity scene in which all of the VR interacted got overridden. A big mistake. As of the moment, there is no showcase/ VR experience like we previously had. With presentations coming, this is a vital thing to rectify. All of the assets. 3D models, and skyboxes are still in the hardrive, which is good news. We just have to put it all back together again. Overall, I think that this will serve as a good learning experience, and some safety measures to implement in the future.


0 comments

“Middle School Tour” 4/26

The groups of 8th graders came through the energy lab today in part of their transitional tour through the HPA high school in preparation for their freshman year. So naturally, we shows the students our VR project, to those that we could give the short amount of time. It went well, I think that it was a little overwhelming and very fast for the 8th graders, but I hope they got something out of it. Afterward, work continued in Unity to try and get the new touch controllers working.

0 comments

"Showcases & Fixing" Week of 4/22 Summary

This week was somewhat different in the way that instead of being able to continue working on the complex process of fixing the gallery scan mesh, etc, rather we got to practice some presentation skills for the project and our methods of creating it. It has been somewhat refreshing to actually share our project with more people, it seems to be a very successful selling point for some students.

Next week I plan on continuing the work of perfecting the gallery environment within Unity, and getting together a comprehensive plan for how presentations will work, and whether or not we can have an open showcase for this project.

0 comments

"More Presenting & Controllers" 4/19

The Oculus touch controllers came today, so Zoe and spent some time trying to get all of that working within Unity, which proved to be challenging. Another group of visitors came by, so it was yet another showcase of our work, and a good chance to practice for the upcoming presentations.

0 comments

"Presenting" 4/17

Today I presented top a group of people coming from Kauai and an Italian family that runs a school in the bay area. This took most of the time and was quite interesting. There are more and more people that come for tours at this time of the year, so some days are almost, essentially dedicated to presenting and sharing our work.


0 comments

“Networking & Optimization” Week of 4/8 Summary

This week was a major success in the both the development of the gallery scan point cloud and a major networking connection that might bring some opportunities in the computer science/ software industry to me and Zoe. It seems like a very fortunate time for this to happen, as mt Senior year is soon coming to a close.


In regard to the gallery scan, I was able to show the results to Mr. Bernstein and he is satisfied with the progress so far. There is much optimization to be done for the Gallery Mesh, starting with reducing the geometry. Texturing the reduced mesh will be the step afterward. Finally, there will have to be some high-quality scans done of the Madge Tennent artworks so that they can be better captured and displayed in the VR & WebGl tour.


Next week I plan on continuing the development of the art gallery scan. While we do have some progress that is sufficient for causal display, I would like a polished and fully deploy-able product before summer break.

0 comments

“Networking” Friday 4/8

Today Zoe and I spent some more time learning Cloud Compare and using it on some parts of the point cloud that reality capture can export. Additionally, a former Zynga employee came to tour HPA and the Elab. Our showcase of our project was a major networking success. He knows many people in the game industry, including VR and even his own recent start-up. I am very excited to see what will come of this.


0 comments

“Could Compare” Wednesday 4/5

Today Zoe and I worked on the point cloud export from Reality Capture and used Cloud Compare for some of the editing. There is a pretty steep learning curve with Cloud Compare, but it is a very versatile and useful tool for editing, matching, and extracting primitive geometry from point clouds. This all makes it very applicable and useful for our project, especially for optimization of geometry for better efficiency inside of a game engine.

0 comments

"Mesh Cleaning" 4/3/17

Today Zoe and I worked on cleaning up the mesh inside of Blender. This was the primary task as was planned last week. I think that things are coming together nicely and that we will have even better results for use in Unity very shortly.

0 comments

"Gallery Scan & Art Center VR" Week of 3/31:4/1

This week has been a major milestone in the development cycle for the Gallery Scan and the work-flow in general. Within (exactly) a week, the entirety of the process was complete with some pretty good results. This is by far the quickest iteration of a “product/demo” with VR capabilities that I was able to create. I think that in retrospect, the most time consuming part of this process in actually getting a good set of images with the right equipment, location, and time. The reconstruction to follow was very, very accurate compared to previous tests of the E-lab rooms where there is simply a lot of clutter and objects that are prone to having errors in the Photogrammetric process.

I have to give big thanks to the Issacs Art Center staff for their help in moving things and ensuring that we have an excellent environment to scan. I also give big thanks to Mr. Bernstien for dedicating a huge amount of time to help this come together, including all of the equipment and planning that was involved.

Next week I plan on doing some mesh corrections for a better experience withing the Unity game engine and a more realistic view. I have been thinking of the design of the entire VR experience so there might be some contextual environment changes to come to make the experience more enjoyable.

Depending on the time we have remaining In the week, I plan to delve into some texturing work after we get a nice and simplified model.

Here is an image of the Gallery in Unity:

Click for full-size image


0 comments

"Meshes & Engines" 3/31/17

@page { margin: 0.79in } p { margin-bottom: 0.1in; line-height: 120% }

Today Zoe and I worked on placing the Gallery Mesh into Unity and doing some extra work to correct it. Zoe did some work in Blender for corrections after Unity was able to successfully import and display the Gallery Mesh. So far, this has been a major success and one of the fastest iterations of getting a full scale scan into Unity and hence, VR.

0 comments

"Gallery Image Reconstruction" 3/29/17

@page { margin: 0.79in } p { margin-bottom: 0.1in; line-height: 120% }

The alignment of the images, along with some reconstruction was completed before class today. I spent some time placing Ground Control Points to improve the reconstruction and create a more accurate mesh. I also exported the reconstructed and textured mesh, along with a simplified iteration to be used in Unity, or the Unreal Engine (in the future). In this way, we have a back-up mesh in case the correction rendition turns out to be less accurate (this can happen sometimes).

Here is an image of the reconstructed point cloud:



0 comments