Team Meetings - HephaestusVision/hephaestus GitHub Wiki
Journal of Meetings and Decisions
- 2012-08-29 Wed - Hal and Chris met with Patrick for initial project meeting
- 2012-08-29 Wed - Hal, Chris, Frank met to organize team.
-
2012-09-05 Wed - Hal, Chris, Frank, Patrick met. (Chris was late to the meeting.)
Before Patrick arrived
- Review differences with C++ and Java
Once Patrick arrived
- Discussed User stories and made changes to third user story
- Discus basics on graphics and polygon options for 3d display
- Discus Midas platform as the server. Wiki is available on midasplatform.org
- Discus using Apache 2 license. It handles patents and is for open and commercial use.
After Patrick left
- Mapped out processes: Kinect - Laptop - cloud - server - cloud - device
- Discussed alternatives and stated that we would look up other options.
- Discuss jobs over the next few days
To Do:
- Hal: Barebones QT + VTK application, put in repository. Check out Midas.
- Chris: Keep working on kinect with PC, and look for open source sdk. Create a progamatic interface (set of functions and classes) for getting the mesh. from the GUI application.
- Frank: Write up functional Spec, and get caught up on Git and C++.
-
2012-09-07 Fri - Hal, Chris, Frank, Prof. Pozefsky
- Talk about adding a comment section to website
- Discuss what is necessary items we need to develop on mac (machine and developer profiles)
- Discuss expanding KiwiViewer for IOS application instead of create one from scratch
- Decide to ask Patrick to get Midas account and make sure it will run on 7/8 year old machine
- Discuss options for technology spike (IOS Development or PCL)
-
2012-09-12 Wed - Hal, Chris, Frank, Patrick
- Talked about getting a Midas server
- Patrick seems happy with the functional spec.
- Discuss applying for jobs at Kitware
-
2012-09-14 Fri - Hal, Chris, Frank, Prof. Pozefsky
- Discussed Mac situation. There is one mac at the library. Swain hall has mac computer lab, but need to get Xcode installed.
- Getting 3 IOS developer licenses under the name Hephaestus
- Discuss VPN server for Midas
- Updated platform selection
After meeting with Pozefsky
- Created schedule for the project.
- Discuss what changes would be applied to KiwiViewer
-
2012-09-21 Fri -Chris, Frank, Prof. Pozefsky
- The department set up a blade for Midas, the computer is a 64bit Ubuntu computer and does not have much memory. If this is a problem we need to speak with Pozefsky.
- There is a Mac our group (any other groups) can use in Sitterson 234.
- Discussed architecture diagram.
- Decided that one technology spike was getting Chris getting the kinect to work with the computer.
- A good option for the next spike would be creating a n IOS app that can communicate with a server.
- Discussed the possibility of having a clean break between the OpenCV model and the VTK file to possible be able to switch to a different file format (from OpenCV to another format).
- Decided to get rid of 3rd user story
- Need to organize schedule in a table.
-
2012-09-26 Weds - Chris, Hal, Frank, Patrick
- Showed Patrick VTK file of Chris' laptop station
- Patrick recommended having a way for the computer app to tell the user when they are done collecting point. For example, when the points match the original points a pop up appears telling the user that they are done.
- Chris needs to check the cloudy class and camera class to make sure they make sense and can be used by kinect data
- Hal will be gone from October 14-21 (the same week as Fall break)
-
2012-09-28 Fri - Chris, Hal, Frank, Prof. Pozefsky
- Mac in 234 is available now. They are putting the newest xcode on it
- If need help with mac, email [email protected] and cc Bill Hayes
- We will present Midas for technology talk on Oct. 31
- On the computer app, we should have a label on each side of the window. One saying model and the other saying camera.
- Do we want the model and camera views to be equal sizes? Way to make middle bar dragable with changeable real-estate.
- Email pointers to cmake and git to Billy Hayes.
- 2012-10-3 Weds - Chris, Hal, Frank, Patrick
- Showed the demo to Patrick
- Contemplated re-sizing the live camera to make it better reveal show what the kinect sees. (maybe make it wider)
- Patrick recommended to reset the camera on start-up of the computer application
- Patrick showed us another website for Midas. midas3.kitware.com/midas
- set up git bug tracking
- 2012-10-5 Fri - Prof. Pozefsky, Chris, Hal, Frank
- Discuss where we are with Midas
- Showed Pozefsky bug tracker on git hub
- Explained color problem with Kinect camera
- Go over schedule
- Showed Pozefsky Midas page dpblo4a.cs.unc.edu (not positive on this website being accurate).
- Registered names on Midas
- any Ipad issue talk to Ipad team, spellgate
- 2012-10-10 Wed - Hal, Frank, Patrick
- Got help with Midas from Patrick
- Told to use midas api
- Midas can return replies in json format
- Need to turn on API module
- 2012-10-12 Fri - Hal, Frank
- Explained how to get/post json replies in QT using signals and slots
- found API methods needed to download items, find folders, and show communities
- 2012-10-24Wed - Hal, Frank, Chris, Patrick
- Patrick showed us a new Kiwiviewer that can access Kitware's Midas server
- Patrick explained that Midas allows but Post and Put. Post has a limit on the size that you can upload an item to Midas. Put does not have this limit.
- Patrick also mentioned that he gave an online course for Midas.
- 2012-10-26 Fri - Hal, Frank, Chris, Prof. Pozefsky
- Told Pozefsky about MidasKiwi that Kitware created.
- Told Pozefsky that the blade went down over fall break, the blade should reset correctly on startup now.
- Told Pozefsky there is no way to create an account in the app. You need to create an account on the website.
- Told Pozefsky where everyone is.
- Frank is trying to build MidasKiwi in xcode.
- Chris is working on merging 2 point clouds
- Hal is working on talking to midas through the scanner application
- Chris explained how his algorithms are supposed to work to merge two point clouds
- 2012-10-31 Wed - Hal, Frank, Patrick
- Patrick explained that we can use cpack (which comes with cmake) to create installer for the scanner app.
- Went to the recycling/printing room to try and build MidasKiwi on the mac.
- Patrick said to email the VES mailing list to get further help to build MidasKiwi.
- 2012-11-2 Fri - Hal, Chris, Frank, Prof. Pozefsky
- Where is everyone at? - Chris working on making the merging of point clouds more efficient - Hal working on finishing touches to make scanner application save a point cloud to the midas server - Frank working on getting the iOS app ready to show for Demo 2 - Pozefsky sent another email to find out about getting iOS developer keys. - Talked a little about Midas presentation - Where does it fit in the world? - What does it do for us?
- 2012-11-7 Wed - Hal, Frank, Patrick, (Chris late)
- Show Patrick Demo 2. - We showed Patrick how to use the scanner app to save pictures to our Midas server - We showed Patrick the code to upload to Midas - We showed Patrick the iOS application to download point clouds that were saved on Midas
- Discussed forking code from iOS app to our git page.
- 2012-11-9 Fri - Hal, Frank, Chris
- Discussed what we need for the User Manual
- Looked at VES source code to check the format that vtk files need to be in to display real color
- To do list
- Hal - Send screenshots of the scanner application to Frank so he can write parts of the User document. Write up document installation. Look into RGB colorArray in VES
- Chris - Create a better version of merge point cloud
- Frank - Write up the rest of the User document. Email Prof Pozefsky on Monday about steps to get iOS app on other peoples devices.
- 2012-11-14 Wed - Hal, Frank, Chris, Patrick
- Patrick is going to compile the scanner application and give some feedback
- We showed Patrick the installation guide
- 2012-11-16 Fri - Hal, Frank, Chris, Prof. Pozefsky
- Show Hal's changes to the Hephaestus viewer computer app
- Explain the algorithm used to combine point clouds
- Brought up question if it is possible to just throw away half the points each time
- Chris will try and use a more sophisticated algorithm to downsample
- Final presentation noon Dec. 10th.
- 2012-11-30 Fri - Hal, Frank, Chris, Prof. Pozefsky
- Show Prof. Pozefsky the ipad app
- Go down to the graphics lab
- Show Prof. Pozefsky progress so far with algorithm
- Discuss possible reasons why the algorithm is failing
- Discuss schedule to finish.