Monday, July 23, 2012

Project Dance Controller: Report 8


This post is part of Project Dance Controller.

It's a bachelor thesis project with the aim of letting the quantity of dance movements in the room control the volume level using the Kinect for Windows hardware.

You can read all reports here.

Last week

Nothing but report writing. I have managed to get the first three chapters done:

  1. Introduction
    1. Background
    2. Problem description
    3. Related research
  2. Application Programming Interface
  3. Implementation
    1. Device detection
    2. Depth data
    3. Motion analysis
Left now is the chapters that describe integration with Stoffi, results and conclusions.

I have also added some more references, now up to 15 papers, articles and books.

Challenges

Finding time and usual but getting the correct wording is hard. I will iterate over the text when it's done and try to improve on some areas where the phrasing might be a bit too informal. It's always good to look at something twice.

This week

I will have the report finished by Friday and use the weekend to polish everything up a bit.

Challenges

Time, time and time. But lack of that hasn't stopped me before. :)

Monday, July 16, 2012

Project Dance Controller: Report 7

This post is part of Project Dance Controller.

It's a bachelor thesis project with the aim of letting the quantity of dance movements in the room control the volume level using the Kinect for Windows hardware.

You can read all reports here.

Last week

Last week I was on a trip to southern Sweden. So that's why there was no report last Monday. This report will thus be about the week before that.

I managed to get the plugin ready and connected to the volume of Stoffi. I have tested the plugin with 1-5 people and it is working pretty well. There's an issue when the computer goes into hibernation and comes back. This will for some reason cause the volume to stagnate around 4 (of max 10). I have no idea why and will not dive into investigating this issue before I hand in the work.

I did not fully implement the new "Filter" type plugin. I have created hooks to allow the plugin to specify various distortions and manipulations such as echo, pitch, etc. I have, however, not connected these hooks to the audio stream.

Challenges

I must admit that I am a bit chocked that the work has progressed so smoothly so far. When the plugin system was extended to allow plugins to alter the volume I could create a plugin and use the DanceAnalyzer library in one day. The majority of the time was spent trying to find good default value for sensitivity and viscosity (resistance to change).

This week

I will now start to write the report. This is what will happen both this week and the next. I aim at having a least more than half but preferably all of the report done by Sunday. That way I can spend most of next week polishing the report, tying up all the documentation, and package the library.

Challenges

Same as always: trying to find enough time. Having a 5 month child is a lot of work but I remain optimistic. I have two weeks left, the report has already been started and I have weekly reports which I can use.

Monday, July 2, 2012

Project Dance Controller: Report 6

This post is part of Project Dance Controller.

It's a bachelor thesis project with the aim of letting the quantity of dance movements in the room control the volume level using the Kinect for Windows hardware.

You can read all reports here.

Last week

I have now extended the plugin platform of Stoffi to allow for a new kind of plugins called "Filters" which essentially are allowed to control and modify various aspects of the audio stream. I have specified the plugin base so it can add various distortions and manipulations but only implemented the actual volume modification into the audio stream itself.

I have added plugin labels so this plugin can show the status of the Kinect device. I also added a new type of setting: slider. It is created by adding a double or integer setting and setting a maximum value. I use this setting to allow the user to specify between which levels of volume that the plugin should operate.

Oh, and I have named the plugin "Shake It!".

Here's a screenshot of the new plugin management panel along with my current skeleton Shake It! plugin (named "Skaka Rumpa!" in Swedish).


I have pushed the documentation writing into next week. I will go on a trip to Skåne then and the car ride will be about 10 hours so writing documentation and the report will be the perfect activity in the car.

Challenges

As usual the biggest challenge has been finding time. The technical part has been flowing very efficiently and great. Not a lot of problems actually.

This week

I will do my best to finish the plugin and incorporate the dance analyzer library into the plugin this week. When that is done all that is left is writing the report and filling out some of the documentation on the wiki.

Challenges

Time, as usual. But I am actually doing pretty well and haven't fallen behind schedule yet so getting these last four weeks together will hopefully not be much of a problem.