For questions about convolutional neural networks, also known as CNN or ConvNet.
0
votes
0answers
18 views
Invalid Argument Error when running simple Convolutional Neural Network [migrated]
I am running a convolutional neural network with CSV files as training and test input. I am getting a strange error that I cannot solve.
...
4
votes
1answer
50 views
Training a CNN from scratch over COCO dataset
I am using Tensorflow Object Detection API for training a CNN from scratch on COCO dataset. I need to use this specific configuration.
There is no pre-trained model on COCO with that configuration and ...
3
votes
1answer
40 views
How to train a CNN
When it comes to CNNs, I don't understand 2 things in the training process:
How do I pass the error back when there are pooling layers between the convolutional layers?
And if I know how it's done, ...
3
votes
0answers
22 views
How should continuous action/gesture recognition be performed differently than isolated action recognition
I am going to train a deep learning model to classify hand gestures in video. Since the person will be taking up nearly the entire width/height of the video and I will be classifying what hand gesture ...
3
votes
2answers
64 views
What is the best approach for writing a program to identify objects in a picture then crop them a specific way?
My works quality control department is responsible for taking pictures of our products at various phases through our QC process and currently the process goes:
Take picture of product
Crop the ...
4
votes
2answers
53 views
Image Classification
I am currently working on a project to classify snake types separately using an image of the snake. I need to train a module to classify snake images, but the problem is there are only a small number ...
2
votes
1answer
45 views
Double pooling layers
In what scenario when assembling a DL CNN would you want to have two adjacent pooling layers, without a convolutional layer between?
0
votes
0answers
16 views
input shape of dataset in CNN [migrated]
My dataset is a simple table of 20 columns and 100,000 rows.It is not a image data as commonly used in CNN. What input shape should I provide in this case?
Right now I did-
...
3
votes
1answer
51 views
Optimizing Max Pooling Algorithm
The below code is a max pooling algorithm being used in a CNN. The issue I've been facing is that it is offaly slow given a high number of feature maps. The reason for its slowness is quite obvious-- ...
2
votes
1answer
53 views
Disentangled VAE doesn't reconstruct accurate grids
I am trying to implement the disentangled VAE model according to this link.
I want to understand the architecture of this model in order to customize it later. As infrastructure, I have a linux kernel ...
0
votes
0answers
5 views
Implementing spatio-temporal convolutions in pytorch [migrated]
I am trying to implement a layer to perform the (2+1)D convolutions described in this paper: https://arxiv.org/pdf/1711.11248.pdf
The basic idea is as follows: Let's say I have a 3D convolutional ...
1
vote
0answers
51 views
Backpropagation of convolutional neural network - confusion [closed]
I've already seen many articles about this topic and Backpropagation In Convolutional Neural Networks by Jefkine (5 September 2016) seems to be the best. Although, as author said,
For the purposes ...
4
votes
2answers
40 views
Loss function for singular object detection
What loss function should one use, knowing that input image contains exactly one target object?
I am currently using MSE to predict center of ROI coordinates and it's width and height. All values ...
6
votes
6answers
317 views
Why Python not C?
I like the enforced indentation of Python that many don't like because I hate parenthetic typing and redundant semicolons. I like the shell interface, but why do some think Python is de facto for ...
3
votes
1answer
62 views
Identifying car model via deep learing
Is there any project or example for a software identifying cars?
Situation: I got multiple angle shots in high resolution from a car. I want the algorithm to tell me "This is a Mercedes SLK" or "This ...
1
vote
0answers
18 views
In CNN (Convolutional Neural Network), does the combination of previous layer's filters make next layer's filters?
I know that the first layer uses a low-level filter to see the edge information. As the layer gets deeper, it will represent high-level (abstract) information. Is it because the combinations of ...
1
vote
0answers
37 views
Using CNN LSTMs for prediction of images from image series
I have the following setup for a prediction task: I want to predict entire pictures from previously given pictures. In my case, only 2 pixels in every frame are neither black nor white, they are some ...
1
vote
0answers
32 views
How do GAN's generator actually work?
I have implemented DCGAN's myself and have been studying GAN's for over a month now. Now I am implementing the pggans but I encountered a sentence
When we measure the distance between
the ...
1
vote
1answer
57 views
How to label training data for YOLO
I am having a question on how to label training data for YOLO algorithm.
Let's say that each label Y, we need to specify [Pc, bx, by, bh, bw], where Pc is the indicator for presence(1=present, 0=not ...
8
votes
4answers
143 views
Beyond neural networks?
Are there possible algorithms that have the potential to replace neural nets in the near future?
And do we need that?
What is the worst thing of using neural networks in terms of efficiency?
2
votes
1answer
31 views
Are there any Neural network benchmarks(forward-pass speed) around?
I have a task where I would like to use a CNN. I would like to incrementally start from the fastest models, fine-tune and see whether they fit my "budget". At the moment, I'm just looking at object ...
3
votes
5answers
165 views
CNN with OpenCV
I have practiced building cnn for image classification with tensorflow, luckily to me they have very good library documentation and tutorials. But i found that tensorflow is too complicated, building ...
2
votes
1answer
65 views
Regarding Yolo and Keras
I'm trying to implement YOLO (tiny version, v1) into Keras framework. For the past two days, I've been relentlessly digging through Github and the likes in order to ...
2
votes
1answer
30 views
Neural network architecture for line orientation prediction
Imagine that a line divides an image in two regions which (slightly) differ in terms of texture and color. It is not a perfect, artificial line but rather a thin transition zone. I want to build a ...
1
vote
0answers
30 views
What are some important results regarding the complexity of training neural networks?
In the paper "Provable bounds for learning some deep representations", an autoencoder like model is constructed with discrete weights and several results are proven using some random-graph theory, but ...
0
votes
1answer
37 views
Using CNN to identify buildings from aerial images
I want to train a CNN (Vggnet) to identify different types of buildings from aerial images.
However seeing that a CNN "ignores" size, e.g. the same type of dog in one image can be large and small in ...
2
votes
1answer
38 views
CNN's vs Densely Connected NN's
In image classification we are generally told the main reason of using CNN's is that densely connected NN's cannot handle so many parameters (10 ^ 6 for a 1000 * 1000 image). My question is, is there ...
1
vote
2answers
38 views
Convolutional neural nets and reduction of the layers
I have a very simple question about Conv nets. I understand the whole principle, but only one thing is not well explained on the Internet.
If I have a 16 channels image that goes on a convolutional ...
4
votes
3answers
65 views
What kind of neural network architecture do I use to classify images into one hundred thousand classes?
I have an image dataset where objects may belong to one of the hundred thousand classes. I want to know what kind of neural network architecture should I use in order to achieve this.
2
votes
2answers
63 views
What is the purpose of “reshaping it into the shape the network expects and scaling it so that all values are in the [0, 1] interval.”?
I am a deep learning beginner recently reading this book "Deep learning with Python", the example explains the process of implementing a greyscale image classification using MNIST in keras, in the ...
1
vote
1answer
136 views
How to load an image into tensorflow.js code which reads handwritten numbers and clasify them
I'm new to machine learning, so i figured I should look into google's tensor flow guides and I know how to code in JS so that's why I'm using tensorflow.js, there's and example in the guide that ...
1
vote
2answers
77 views
Speeding up CNN training
So I built a CNN without any scientific libraries like TensorFlow or Keras (only NumPy). It is taking a huge amount of time to train. What are some of the tricks and tips followed by people to speed ...
0
votes
0answers
16 views
Building a VAE architecture to reconstruct images
I am trying to build a VAE architecture in Lasagne which is able to reconstruct MNIST images. Note that I want to be able to sample from both the encoder and the decoder.
So far, my architecture is ...
1
vote
0answers
70 views
Object recognition by two or more traits that are orthogonal (informally speaking)
I would really appreciate if someone could comment the following method of training neural nets providing them with some meta data (Making them more color prone only if needed, whereas now they're ...
1
vote
3answers
73 views
Which photo is more artistic?
I would like to develop a machine learning algorithm, given two photos, that can decide which image is more "artistic".
I am thinking about somehow combining two images, giving it to a CNN, and get ...
1
vote
2answers
42 views
How to make a fair comparison of a convolutional neural network (cNN) vs a mutlilayer perceptron (MLP)?
I'm working with deep learning on some EEG data for classification, and I was wondering if there's any systematic/mathematical way to define the architecture of the networks, in order to compare their ...
3
votes
0answers
37 views
Relative compute time for each type of layer in a neural network
Hello,
I would like to know whether this picture from the paper: Distributed Training of Deep Neural Networks: Theoretical and Practical Limits of Parallel Scalability valid?
Questions:
1) Does ...
3
votes
1answer
76 views
What is feature embedding in the context of Convolutional Neural Networks?
What are feature embeddings in the context of Convolutional Neural Networks? Is it related to bottleneck features or feature vectors?
1
vote
1answer
22 views
Detecting Keypoint of 3D model, and distance between them
I am very new to AI, I have a set of 3D human models that I would like to train the algorithm to identify wrist, upper arm, lower arms, etc, and distance between them.
From my understanding, this is ...
2
votes
1answer
181 views
Detecting license plate using tensorflow
I'm currently working on license plate recognition. My system consist of 2 stage: (1) License Plate region extraction & (2) License Plate region recognition.
I'm doing (1) with Raspberry pi 3 ...
1
vote
1answer
112 views
Custom Object Detection model including orientation of the Object
I am looking to detect think objects like pens, pencils and surgical instruments. The bounding box is not important, but I am looking to see if I can train a model to detect both the object as well as ...
5
votes
2answers
52 views
Can translational invariance of CNNs be unwanted if object is likely in certain positions?
Various texts on using CNNs for object detection in images talk about how their translation invariance is a good thing. Which makes sense for tasks where the object could be anywhere in the image. Let'...
1
vote
0answers
37 views
Normalizing height data for CNN
A task I’m working on at the moment requires a CNN with a height map as one of the inputs. This is a matrix of floating point values in which each point is the height of that point above sea level.
I’...
2
votes
1answer
95 views
Why to prefer ReLU over Linear activation functions?
ReLU : y = max(0,x)
Linear : y = x
The ReLU nonlinearity just clips the values ...
1
vote
0answers
15 views
Atrous (Dilated) Convolution: How one can compute responses of arbitrarily high dimensions in DCNN?
According to this paper (page 4, bottom-right), atrous convolutions can be used to compute responses of arbitrarily large dimensions in Deep Convolutional Neural Networks.
I do not understand how ...
1
vote
1answer
74 views
Is this overfitting avoidable?
I am trying a modification of Mobilenet in which I add feedback from the softmax layer into the early layers (to implement this I put a second net after the first, which receives connections from the ...
1
vote
0answers
33 views
Capsule Networks - Facial Expression Recognition
I want to experiment Capsule Networks on FER. For now I am using fer2013 Kaggle dataset.
One thing that I didn't understand in Capsule Net was in the first conv layer, size was reduced to 20x20 - ...
3
votes
2answers
239 views
Convolutional neural networks with input images of different dimensions - Image segmentation
I'm facing the problem of having images of different dimensions as inputs in a segmentation task. Note that the images do not even have the same aspect ratio.
One common approach that I found in ...
3
votes
2answers
59 views
VAE image reconstruction question?
Is it possible to use a VAE to reconstruct and image starting from an initial image instead of using K.random_normal as show in the “sampling” function of this example?
I have used a sample image ...
0
votes
0answers
6 views
how to process the two different size of feature map (from VGG) into the same size?
I have two convolutional neural networks outputting two feature maps which is [1,1000] from VGG and [1,3852], I need to have a process to make the two feature with the same size of feature, for ...
