46
On the top of this answer, you can see a section of Updated links, where artificial intelligence, deep learning or and database machine learning progressively step of the grounds of traditional signal processing/image analysis/computer vision. Below, variations on the original answer.
For a short version: successes of convolutional neural networks and deep ...
21
First, there is nothing wrong with doing grad work in image processing or computer vision and using deep learning. Deep learning is not killing image processing and computer vision, it is merely the current hot research topic in those fields.
Second, deep learning is primarily used in object category recognition. But that is only one of many areas of ...
15
By long shot it is doable - to what extend? You will see. This task of environmental sound classification is not very well studied. Also choice of machine learning paradigm is crucial - statistical approach or maybe binary classifier? You can start with GMM's, ANN's and SVM's - I opt for GMM's and ANN's.
Yes, most of people are using MFCC's because they are ...
11
No Deep Learning isn't killing Image Processing. You need huge datasets and lots of computational resources to do deep learning. There are plenty of applications where it is desirable to be able to do image processing with less computational load and smaller memory footprints and without having access to huge databases. Some examples are mobile phones, ...
11
Today we had a discussion with a friend of mine. It was a rainy day here in Munich, while a large portion of Europe was having a kind of sunny atmosphere. People were sharing photographs in social media, where they were in nice summer dresses, wandering around the seas. She was annoyed with this situation and turned to me and asked: "Could you write a ...
8
Some Features:
Mean.
Variance.
Skewness.
Kurtosis.
Dominant 3 frequencies in the DFT.
Energy of the 3 dominant frequencies.
Max Value.
Min Value.
Median.
Usually I'd compute them in running windows.
Another great information is the Histogram of the Derivative.
Or just all the above of the Derivative.
7
The short answer is, No. DL can recognize a mug in a photo, but this doesn't kill signal processing in anyway. That said, your question is quite relevant in these troubled days. There is a nice panel discussion on the subject, featuring Stephane Mallat, etc., here.
5
It can be done very easily with the scikit-learn. Examples are easy to find on their website, i.e. here. In my opinion it is the best way to go.
Modified code example from the above link:
import numpy as np
import matplotlib.pyplot as plt
from sklearn.cluster import KMeans
from sklearn.datasets.samples_generator import make_blobs
########################...
5
In addition to the features mentioned so far I would like to mention measures of complexity such as:
Shannon Entropy
LZ Complexity
Fractal Dimension
There are also Fourier Descriptors (as hinted by Drazick already) and their equivalent in Wavelet Analysis and of course simple histogram bins which would return how frequently each gear is engaged en route.
...
5
Data engineering is still used in machine learning to preprocess and select the data fed to DNNs to improve their learning time and their evaluation efficiency. Image processing (the stuff between the camera sensor and the RGB/etc. bitmaps fed to the DNNs), a form of data engineering, is still needed.
5
There are, a few discrepancies that might be making a difference here. My suggestion would be to edit the question for clarity. There are quite a few assumptions that lead to non-straightforward thinking about the problem which I have tried to address to an extent and I would be happy to modify the response in light of more information.
In machine ...
5
From feature extraction to learning the desired result, deep learning algorithms can act as full pipelines for solving tasks at hand. End-to-end learning usually refers to omitting any hand-crafted intermediary algorithms and directly learning the solution of a given problem from the sampled dataset. This could involve concatenation of different networks ...
4
Non-verbal Audio (let alone environmental) seems to be the little brother to main stream machine learning media types like images, speech, text.
To answer your question is it possible to train a network to identify a given sound? Yes, it is! But it's hard for all the same reasons machine learning is hard.
However what's really holding Audio back, and why ...
4
A thorough understanding of signal processing (along with linear algebra, vector calculus, mathematical statistics etc.) is imo indispensable for non-trivial work in the field of deep learning, especially in computer vision.
Some of the high impact papers in deep learning (now that most of the low hanging fruit have been picked) evince a good understanding ...
4
Types of signals:
According to their range set (values): Real Valued, Complex valued ;
According to their dimensions: Scalar, Vector ;
According to their values: Continuous Amplitude, Quantized ;
According to their domain set (arguments): Continuous-time, Discrete-time :
According to their mappings: Deterministic, Stochastic (Random) ;
According to their ...
3
Labelling a missing value as -1000 (assuming a non-missing value is much smaller in absolute value) will not cause problems with anything based on decision trees/stumps (bagged/boosted). You might also have some success with mixture models, for example a GMM - the learning algorithm will allocate some components to cover the missing data. Same for non-...
3
Here is a solution for sound classification for 10 classes: dog barking, car horn, children playing etc. It is based on tensorflow library using neural networks. Features are extracted by converting sound clips to spectrogram
3
The idea underlying the k-means algorithm is to find clusters that minimize the intra-cluster variance and maximize the inter-cluster sum-of-square. Note that the total variance is fixed so minimization of intra-cluster variance is equivalent to maximization of inter-cluster sum of square. Basically various methods aim to achieve such minimization.
However, ...
3
What you are asking is called 'page layout analysis' in OCR. Take a look at https://en.wikipedia.org/wiki/Document_layout_analysis.
3
My perspective from university was that many signal processing people were a bit hostile toward ML, I suspect because they felt threatened that it was encroaching on their domain. But recently there's been a lot of research into benefits of complex valued deep neural networks, which may suggest that the golden ticket is really a solid understanding of both ...
3
This is a concept in supervised machine learning.
Train data: Used to train your supervised ML model. This data contains both the input and the desired output, which is compared with the output from your model. Your model is hence trained to reduce the prediction error or cost function.
Test data: The data for which you want to predict the outcome. The ...
3
Is this necessary if I only intend to use the data in the Neural Network toolkit provided by the repo I linked?
Yes.
Whether or not you are downsampling (instead of just decimating) has nothing to do with classification performance but rather, it is to preserve (as much as possible) the information contained in the signal.
When changing the sample rate ...
3
Is there a script / tutorial / demo for penis detection?
[...]
Fairly serious quesion, future of internet memes is at stake
Yes, there is.
Common Pattern Recognition techniques will be able to spot one even with what would be considered today "traditional" approach (i.e. without "Deep Learning").
There already is a sub-category of the subject ...
3
Band-pass filtering with cut-off frequencies of 300 Hz and 3400 Hz should result in a good approximation. Try with a Chebychev filter or order not more than 6.
Then you may need to downsample your audio to 8000 samples per second, which is the standard for telephony.
P.S. The actual cut-off frequencies (especially the 3400 Hz) may be different according to ...
3
General Idea
The general idea of Principal Component Analysis (PCA) is as following (Intuition over formalism):
Given a set of points in space (Inner Product Space) find a set of vectors (Directions) which are uncorrelated which span the data in the most energy preserving manner.
The tricky part is explaining "most energy preserving manner".
So we're ...
2
I would use the SVD (Singular Value Decomposition).
By looking at the Singular Values I'd determine which vectors spread the data and which spread the noise.
Practically, they both do both, but if we speak which are dominant, this would be a great starting point.
2
At your first stage, which classifier to use is not that important, you may need to make your code run first. There was a very famous paper on Face Recognition and Gender Determination, in which a very complex transform was implemented to represent each node on the face. The used gabor kernal based wavelet transform to get set of coefficients for each node(...
2
I think you've got the point, but that you are getting confused by the additional tables that you think should be used to implement buckets.
Let's rephrase the quote from the tutorial first:
In LSH, you create a partition of the entire space. If you use a large
bucket width $w$, then obviously you need less buckets to span the
entire space and thus ...
Only top voted, non community-wiki answers of a minimum length are eligible
