Questions tagged [neural-network]
The neural-network tag has no usage guidance.
73
questions
2
votes
1
answer
34
views
2d convolution: What is the difference between convolution using blocked Toeplitz matrix and convolution layers?
For a given matrices $A$ of size $4\times 4$ and $B$ of size $3\times 3$ then I construct a blocked Toeplitz matrix and perform the convolution. The resulting output is of size $6 \times 6$.
I have no ...
- 655
1
vote
0
answers
46
views
Processing of Channel State Information from WiFi for Generalizability
For my master's thesis I collected CSI with 4 WiFi network cards (Intel AX210) for three different persons for 102 positions (spaced roughly 30cm apart).
The covered area was around 10m x 7m and in ...
- 111
1
vote
1
answer
64
views
Tapped delay line + ADALINE = Adaptive filter?
When studying neural networks from Neural Networks and Learning Machines, by Simon Haykin, the author highlights the close similarity between of adaptive filtering and neural networks.
From a scalar-...
- 337
1
vote
1
answer
59
views
preparing the ct scan data of a patient, using a visible feature in the mri image
A dataset of mri and ct scan images of patients has been prepared. There is a feature /damaged area/ in the mri image that is easily visible. But the injury of this area is not visible in the CT scan ...
- 11
4
votes
1
answer
124
views
Understanding an adaptative single neuron PID controller
I only know the "vanilla" use of a Kalman filter and I am currently trying to understand an article available here (the algorithm is presented in the 6 first pages) :
Adaptive Single Neuron ...
- 487
1
vote
0
answers
101
views
State-of-the-art non-training based stereo matching algorithm
In the world of 3D vision, constructing a 3D model from multiple view imagery is an important topic. The stereo matching step is one of the most crucial steps. Based on the benchmark of KITTI2012 and ...
- 213
1
vote
1
answer
126
views
Log of Filterbank Energies
In common literature, when generating spectrograms, mel-spectrograms, and cochleagrams, the log of the resulting filterbank energies is taken. Why is this done? I notice that my convolutional neural ...
- 11
0
votes
0
answers
22
views
Question about "A neural algorithm of artistic style"
My question is based on the paper. This paper introduced a loss function by calculating two-loss: the loss between the white noise input and style image and the loss between the white noise input and ...
- 23
0
votes
1
answer
241
views
How to convert depth to disparity
if $k = \text{baseline}\cdot\text{focal length}$ is known, then the disparity is the ratio of $k$ to depth $d_\text{image}$:
$$D_\text{image} = \frac{k}{d_\text{image}}$$
I have a ...
- 19
0
votes
1
answer
586
views
normalize STFT output by magnitude
I am using torch.stft() to generate spectrograms for neural networks and come across the below code.
...
- 55
3
votes
2
answers
320
views
Optimize window length (STFT) via gradient descent (in neural networks)
The authors from this paper optimized a Gaussian window size via gradient descent (the σ parameter of the bell curve) together with the other parameters of neural networks.
I don't use Gaussian window ...
- 55
0
votes
0
answers
28
views
How can I speech enhancement with DNN?
I want to do speech enhancement with DNN on Matlab. I downloaded the TIMIT dataset from the internet as a .wav file. I also downloaded the noises from the Noisex-92 database as .m file. Now I need to ...
- 59
2
votes
2
answers
546
views
Is a neural network an adaptive filter?
I am confused as to the difference between neural networks and adaptive filters: As far as I understand it, "neural networks" are largely used for solving inverse problems, where an unknown ...
- 207
5
votes
2
answers
477
views
Explain the Process of Spectral Pooling and Spectral Activation in the Context of CNN in Frequency Domain
I am reading the paper Design of an energy efficient accelerator for training of convolutional neural networks using frequency Domain Computation:
which uses Frequency Pooling, from Spectral ...
- 353
4
votes
1
answer
243
views
Perform Transposed Convolution in Spectral / Frequency Domain?
I'm doing some experimentation on performing end to end generative modeling in the frequency domain. I've got a working convolutional layer, but do not yet have a Conv2DTranspose equivalent. Please ...
- 43
4
votes
1
answer
76
views
The Meaning of $ \mathbb{E} $ Operator in the Pix2Pix Loss Formula of a Neural Network / Convolutional Neural Network
I've been observing the Pix2Pix Paper - Image to Image Translation with Conditional Adversarial Networks and wondered on formulas. For example, the objective of the CGAN is:
,
where x - observed image,...
- 173
0
votes
1
answer
24
views
Training a NN classifier on a single\double channel out of a surround sound dataset
I want to train a neural network for a classification task on the VSD2014 dataset. I have downloaded the movies, but they have a 6 channel audio format (surround sound). 6 channels will cost a very ...
- 672
1
vote
1
answer
669
views
Using STFT as an input to a Neural Net
I'm trying to use the STFT as an input to a neural network. After flattening, there are over 4,000 features for a few seconds of audio. Is there a recommended way of summarising these to be a more ...
- 141
0
votes
0
answers
73
views
Filter Bank and Auto-Encoder
I'm trying to find an intuition behind auto-encoder using an analogy with filter banks.
I can comprehend the encoder and analysis filters in a filter bank as extracting features from the input signal ...
- 101
1
vote
1
answer
495
views
a neural network approach for FIR filter
I am trying to write a code for a neural network to do the digital filtering on some signals. Is there any neural network model for digital filtering?
- 11
2
votes
1
answer
5k
views
Upsampling vs downsampling. Which to use when?
Downsampling reduces dimensionality of the features while losing some information. It saves computation.
Upsampling brings back the resolution to the resolution of previous layer.
My question is which ...
- 21
0
votes
0
answers
37
views
How can I find the delay between two signals? [duplicate]
I am using NARX in Matlab.
Is there any method to find the delay between input and output signals?
My aim is to decrease memory length by finding the delay.
TIA.
- 1
0
votes
0
answers
26
views
How can a CNN account for spectro-temporal constraints in neural data?
What are there the best ways to leverage the unique "geometrical" constraints of spectro-temporal signal representations (architecture, filter shapes, data augmentation, etc.)? For example, ...
- 1
1
vote
1
answer
41
views
FFT question as it relates to Neural Networks / Supervised Learning Models
So I've been researching wavelet transforms and FFT. I want to feed wavelet transforms of a 1D signal in time into a neural net and train against a target variable at each time step. The idea being ...
- 11
1
vote
1
answer
325
views
How to train and test deep neural network using MFCC features?
I am working on Voice Disorder Detection problem. I have extracted 13 MFCC features, 13 delta and 13 delta-delta features from each audio file (2 to 4 secs). I extracted these features for each frame (...
- 11
3
votes
2
answers
164
views
How to Design a Model based on Convolutional Neural Network (CNN) which Supports Arbitrary Input Size in Training and Production
I'm doing a research about personality identification based on their signature using CNN method, however the learning feature for the personality traits have a different input size. I understand that ...
- 141
0
votes
0
answers
21
views
Functioning of a continuous data stream processing neural network?
I've been searching with unfortunately no relevant results about something I have trouble to figure out: neural networks which process continuous data stream (audio, video, anything with a state ...
-1
votes
1
answer
406
views
STFT to spectrogram
I would like to know whether I am correct in my understanding of going from STFT to a spectrogram. My goal is to convert a spectrogram back to a wav file.
If I have my STFT:
...
- 153
1
vote
2
answers
98
views
Neural networks in system identification - What type of activation functions?
I made a free software for all operative systems, even Android. It's called Deeplearning2C. It can train a neural network and generate C code and MATLAB-code. C-code for embedded systems and MATLAB-...
- 403
0
votes
0
answers
98
views
Using unlabeled EEG data for Machine Learning
I am working on a project that is basically a game for motor-paralyzed people.
It should take an EEG signal from FP1 channel of brain and then after processing it should generate command for the game ...
2
votes
1
answer
124
views
Deriving the Langrangian interpolation polynomials in Cook-Toom convolutions
I'm working through Blahut's 'Fast Algorithms for Signal Processing'. Trying to develop an intuition for the Cook-Toom algorithm for convolutions as used by Lavin and Gray in their Winograd paper for ...
- 123
1
vote
1
answer
73
views
Develop a simple image classification system by using Laplacian of Gaussian (LoG)
I am tring to write a simple CAD system to classify some images within two groups by using Laplacian of Gaussian (LoG). I am using scikit-image for this tasks and I want to use DNN in keras to train a ...
- 11
-1
votes
1
answer
81
views
Reconstruct images from PCA reduced dimensions with NN
I was reading this Medium post and I had the idea to reconstruct the original images with a convolutional neural network instead of applying the inverse transform method. The problem is that I don't ...
- 51
0
votes
0
answers
27
views
What features can I extract from optical fibre signal?
I have sensors placed in pipelines and I have amplitudes coming from it. I want to detect the type of activity taking place at the sensors(like digging,tunneling,clamping etc). So I have a signal ...
- 101
0
votes
0
answers
447
views
Upsampling and downsampling signals as a preprocessing step for a neural network
I have audio data acquired from a 4 channels sensors array.
As a preprocessing step for a neural network, I want to beamform and focus on the sound source. For higher resolution in the beamforming ...
- 672
5
votes
4
answers
345
views
After advice about detecting focus quality of objects in a photo detected using YoloV3
I've spent the last couple of days playing with YoloV3, and have had very good results. My use case is sports photography, and the object detection for people/bikes etc is very very good, I'm very ...
- 53
4
votes
1
answer
822
views
How to Remove the Patch Artifacts of Neural Network Denoising Process?
I have written a python script which uses the Noise2Noise: Learning Image Restoration without Clean Data implementation of the Auto Encoder which is useful to remove noise from images. In the original ...
- 155
0
votes
1
answer
112
views
What does it mean for a Wavelet transform to commute with translations?
Referencing this article here https://arxiv.org/pdf/1203.1513.pdf
It states "A wavelet transform commutes with translations, and is therefore not translation invariant". Now I understand why it is a ...
- 793
1
vote
0
answers
112
views
Image from Inverse FFT operation is sparse
I'm trying to train a deep nerual network that takes in input, an signal in the frequency domain, and attempts to learn a mapping to another signal in the frequency domain.
Basically, the input to ...
- 11
1
vote
1
answer
548
views
Generate audio data using VAE+GAN
I am trying to train a VAE+GAN model to generate sounds produced by honeybees. I build my model by slightly modifying this tutorial, which aims to generate new MNIST images. Since my data are 1D ...
- 183
0
votes
0
answers
850
views
Recover Audio from Spectrogram Image
I have applied transformation(Constant Q Transformation) to my audio time domain signal, this gives me Frequency vs Time representation. Now, the CQT values consists of complex values(84*260), but ...
- 101
2
votes
1
answer
2k
views
What is the best input for de-noising autoencoder for sound data?
I am currently trying to build an autoencoder to de-noise audio data.
However I have not found any good articles explaining about the input to the autoencoder, i.e. feature vector.
As in speech ...
- 21
1
vote
1
answer
244
views
Techniques to reject noisy neural network input
Suppose an artificial neural network is used to approximate a sine wave (shown in red in the graph below), given the linear input variable $x$ (scaled such that the ANN input is $x_{\rm nn}\in[-1;1]$)....
- 165
0
votes
1
answer
275
views
Generalized translation on graph
David I.Shuman in "vertex-frequency analysis on graph" claims that,"we generalize one of the most important signal processing tools – windowed Fourier analysis – to the graph setting and When we apply ...
7
votes
1
answer
646
views
"Bi Directional" Kalman Filter - Kalman Filter for Smoothing
I am working on a project in Object Tracking, i.e. need to predict the location of next bounding box.
I used a Hungarian algorithm with a Kalman Filter (which is a common method in this domain) which ...
- 181
4
votes
1
answer
4k
views
Is there a penis-detection demo similar to face-detection?
Is there a script/tutorial/demo for penis-detection similar to this one on face-detection?
This is a fairly serious question, as the future of Internet memes is at stake. Breast / nipple detection ...
- 165
1
vote
1
answer
5k
views
Convolution and Cross Correlation on 2D Image
I have read that convolution and cross-correlation are the same thing, but convolution flips 180 degrees (images), or time reverses (sequences) the kernel, before performing the elementwise ...
- 209
7
votes
3
answers
4k
views
Neural Networks and Complex Valued Inputs
[not sure if this or stats.stackexchange was the correct location for this post, so put it on both for now.]
I've seen some recent papers describing complex valued neural networks like this one: Deep ...
- 281
5
votes
1
answer
3k
views
Downsampling audio for use in Machine Learning
I'm trying to use the work (Neural Networks) done in this repo:
https://github.com/jtkim-kaist/VAD
It says this:
Note: To apply this toolkit to other speech data, the speech data
should be ...
0
votes
1
answer
108
views
Neural Network learning project based on 8 wave signals over 1 second at 1 sample every 10 ms ( hence 100Hz )
I'm currentely trying to train a neural network that can decide wether a pattern produced by the movement of a hand near capacitive sensors is as expected, or random.
I have an MPR121 microchip linked ...
- 3