Comments (3)
Let's consider 3 variables: nLevels, level, and nRefinedLevels.
Variable level means the number of discretization steps of the search space for localization. The number of points on the full sphere corresponds to N = 10 * 4 ^ L + 2, where L is the number of levels.
L = 0, N = 12
L = 1, N = 42
L = 2, N = 162
L = 3, N = 642
L = 4, N = 2562
...
The variable nLevels represents the number of different resolution levels. For instance, we often have nLevels = 2, and L = 2 and L = 4. This means a first coarse scan is performed on a sphere with a resolution L = 2, and then for the most probable region of sound source activity, a fine scan is done in a sphere with L = 4 for more resolution. Having multiple levels allows to reduce computations instead of scanning entirely the full resolution sphere.
Finally, nRefinedLevel is used in the initialization step to discretize the continuous space around each point to optimize the size of the sliding window. My advice is to leave this parameter to 1.
from odas.
Thank you. I will take a deeper look into it.
from odas.
Let's consider 3 variables: nLevels, level, and nRefinedLevels.
Variable level means the number of discretization steps of the search space for localization. The number of points on the full sphere corresponds to N = 10 * 4 ^ L + 2, where L is the number of levels.
L = 0, N = 12 L = 1, N = 42 L = 2, N = 162 L = 3, N = 642 L = 4, N = 2562 ...
The variable nLevels represents the number of different resolution levels. For instance, we often have nLevels = 2, and L = 2 and L = 4. This means a first coarse scan is performed on a sphere with a resolution L = 2, and then for the most probable region of sound source activity, a fine scan is done in a sphere with L = 4 for more resolution. Having multiple levels allows to reduce computations instead of scanning entirely the full resolution sphere.
Finally, nRefinedLevel is used in the initialization step to discretize the continuous space around each point to optimize the size of the sliding window. My advice is to leave this parameter to 1.
Hi, great explanation! Could you also help with giving a bit explanation on this: I am using a minidsp 8-channel microphone for running ODAS, and below is some raw output from terminal - I know the first part is the result of SSL and the second part is the result of SST, the 'x,y,z ' in SSL result indicates the 3d vector of sound sources in the space and the 'E' stands for energy, indicating the intensity of sound.
{
"timeStamp": 2351,
"src": [
{ "x": -0.500, "y": 0.162, "z": 0.851, "E": 0.268 },
{ "x": -0.564, "y": -0.097, "z": 0.820, "E": 0.159 },
{ "x": -0.199, "y": 0.274, "z": 0.941, "E": 0.131 },
{ "x": -0.906, "y": 0.420, "z": 0.059, "E": 0.105 }
]
}
{
"timeStamp": 2351,
"src": [
{ "id": 0, "tag": "", "x": 0.000, "y": 0.000, "z": 0.000, "activity": 0.000 },
{ "id": 34, "tag": "dynamic", "x": -0.458, "y": -0.807, "z": 0.374, "activity": 0.000 },
{ "id": 0, "tag": "", "x": 0.000, "y": 0.000, "z": 0.000, "activity": 0.000 },
{ "id": 0, "tag": "", "x": 0.000, "y": 0.000, "z": 0.000, "activity": 0.000 }
]
}
And I am wondering:
What do the "x,y,z" , the "ID" tag and the 'activity' in the SST result mean?
Why there are only 4 lines of output for each?
Which sound source is tracked in the result of SST? Is it the one with the highest energy?
from odas.
Related Issues (20)
- Working with pseye problem HOT 1
- Sound Source Localization with 3 Respeaker 4 Mic array HOT 2
- Change Timestamp
- angle
- Is there a tutorial/guide on odas code? or could you add more comments?
- Is there a Python wrapper for this library HOT 3
- Trouble with Sound Source Separation
- Decibel Level
- Respeaker USB not working with direct capture HOT 2
- Persistent source tracking?
- odas automatically closing on pi zero w
- The definition of uniform distribution is different between the paper and code HOT 1
- How can I revise the SSL section for seeing some parameters? HOT 1
- Hardware list
- Tunning ODAS for a noisy robot
- sound source separation tuning
- Which configuration file shall I choose, if I have a 8-array microphone? HOT 1
- What does the raw output of ODAS mean?
- What does the raw output mean? HOT 1
- How to change the output frequency?
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from odas.