A message from our CEO about the future of Stack Overflow and Stack Exchange. Read now.

Questions tagged [convolutional-neural-networks]

For questions about convolutional neural networks, also known as CNN or ConvNet.

Filter by
Sorted by
Tagged with
-1
votes
0answers
24 views

How could one track the feature locations in a convolution neural network?

A major problem with deep learning, according to Hinton, is that operations like max-pooling remove the position information of features with respect to each other. How one might attempt to track ...
2
votes
0answers
25 views

CNN for image-to-image mapping

I am working on a problem in which I need to train a neural network to map one or more input images to one or more output images (1 channel for image). Below I report some examples of input&output....
1
vote
0answers
8 views

Scoring feature vector with Support Vector Machine

I am reading the R-CNN paper by Ross Girshick1 et al. (link) and I fail to understand how they do the inference. This is described in the section 2.2.Test-time Detection in the paper. I quote: At ...
1
vote
1answer
14 views

Best Camera and protocol for embedded real time CNN project

I'm looking to develop a stand-alone real-time outdoor imaging CNN application, and I can't wrap my head around the myriad of camera options and their communication protocols. The target is a Linux ...
1
vote
1answer
49 views

Why everyone is using CNN for image segmentation?

I'm newbie in artificial intelligence. I have started to research about how to do image segmentation and all the papers that I have found are about CNN. Most of them use the same network, U-NET, but ...
1
vote
0answers
28 views

Reinforcement learning CNN input weakness

I'm trying to train a network to navigate a 48x48 2D grid, and switch pixels from on to off or off to on. The agent receives a small reward if correct, and small punishment if incorrect pixel plotted. ...
2
votes
0answers
16 views

How many ways are there to perform image segmentation?

I'm new in Artificial Intelligence and I want to do image segmentation. Searching I have found these ways Digital image processing (I have read it in this book: Digital Image Processing, 4th edition)...
0
votes
0answers
37 views

Why am I getting low accuracy for my model to generate CAPTCHAs? [closed]

I have a model for solving captcha trained with a dataset but my training images are different and I get bad accuracy for the target data. I am not able to generate captchas in the style of the target ...
1
vote
0answers
14 views

When training a CNN, what are the hyperparameters to tune first?

I am training a convolutional neural network for object detection. Apart from the learning rate, what are the other hyperparameters that I should tune? And in what order of importance? Besides, I read ...
1
vote
0answers
16 views

Training dataset for convolutional neural network classification - will images captured on the ground be useful for training aerial imagery?

I am an agronomy graduate student looking to classify crops from weeds using convolutional neural networks (CNNs). The basic idea that I am wanting to get into involves separating crops from weeds ...
0
votes
1answer
16 views

Set my own kernels to a CNN and don't let it to modify it

I'm newbie in Convolutional Neural Networks and I have discovered (and I hope I'm right) that kernels in convolutional layers are learned while training. If I have a kernel that it is very good to ...
3
votes
1answer
38 views

How to draw bounding boxes for gender classification?

I wonder what is the better way of drawing rectangles on images for gender classification. My task is to create a classifier (CNN based) to detect gender from pictures of entire bodies (not just faces)...
3
votes
0answers
29 views

What are some ways to quickly evaluate the potential of a given NN architecture?

Main question Is there some way we can leverage general knowledge of how certain hyperparameters affect performance, to very rapidly get some sort of estimate for how good a given architecture could ...
2
votes
1answer
18 views

Semantic Segmentation For Multiple Objects When Trained On Single Object

More of a conceptual question here: I'm working on semantic segmentation tasks in the medical space using the U-Net. Let's say that I train a U-Net model on medical images with the goal of segmenting ...
2
votes
0answers
46 views

Suggestion for finding the stable regions in spiral galaxy data?

I am working with a data set that consists of the actual pitch angle (given as PA(Y)) and the pitch angle at each radii (listed from 1 to 217). In the image below, ...
2
votes
0answers
14 views

Is it normal to see oscillations in tested hyperparameters during bayesian optimisation?

I've been trying out bayesian hyperparameter optimisation (with TPE) on a simple CNN applied to the MNIST handwritten digit dataset. I noticed that over iterations of the optimisation loop, the tested ...
0
votes
1answer
49 views

Relationship between input range and channel means, standard deviations for CNNs

So, I'm using a pretrained pnasnet5large model to do some image classification (https://github.com/Cadene/pretrained-models.pytorch/blob/master/pretrainedmodels/models/pnasnet.py) In the file, it ...
1
vote
0answers
11 views

YOLOv3 Synthetic Data Training

Suppose we want to train a model to detect various objects. Let's say we have training data of those objects in various backgrounds along with their bounding boxes. Basically these objects have been ...
0
votes
0answers
19 views

How to make Hopenet (face pose regressor) to work from github? [closed]

Please add tags to this question, I don't really know the correct ones. I am trying to make Hopenet run on my computer, using the code on github I have installed all required libraries. This is my ...
1
vote
1answer
154 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 - ...
1
vote
1answer
57 views

What is a cascaded convolutional neural network?

For a project I am doing, I found the paper Face Alignment in Full Pose Range: A 3D Total Solution. It is using a cascaded convolutional neural network, but I wasn't able to find the original paper ...
3
votes
0answers
27 views

How can I do hyperparameter optimization for a CNN-LSTM neural network?

I have built a CNN-LSTM neural network with 2 inputs and 2 outputs in Keras. I trained the network with model.fit_generator() (and not ...
4
votes
0answers
28 views

Autoencoder produces repeated artifacts after convergence

As experiment, I have tried using an autoencoder to encode height data from the alps, however the decoded image is very pixellated after training for several hours as show in the image below. This ...
2
votes
2answers
51 views

Calculation of FPS on object detection task

How to calculate mean speed in FPS for an object detection model like YOLOv3 or YOLOv3-Tiny? Different object detection models are often presented on charts like this: I am using the DarkNet ...
3
votes
1answer
84 views

How to formalize learning in terms of information theory?

Consider the following game on a MNIST dataset: There are 60000 images. You can pick any 1000 images and train your Neural Network without access to the rest of images. Your final result is ...
3
votes
1answer
29 views

How are small scale features represented in an Inverse Graphics Network (autoencoder)?

This post refers to Fig. 1 of a paper by Microsoft on their Deep Convolutional Inverse Graphics Network: https://www.microsoft.com/en-us/research/wp-content/uploads/2016/11/kwkt_nips2015.pdf Having ...
3
votes
1answer
53 views

How to create a fully connected(matrix) layer with vector input

I am trying to replace last fully connected layer of size 4096/2048 with a matrix of size 100x300 with previous fc layer output of 2048. I've tried 2D convolution - to map from 2048 --> 100x300 (...
2
votes
1answer
63 views

The reasoning behind the number of filters in the convolution layer

Let's assume an extreme case in which the kernel of the convolution layer takes only values 0 or 1. To capture all possible patterns in input of $C$ number of channels, we need $2^{C*K_H*K_W}$ filters,...
2
votes
0answers
13 views

Which CNN hyper-parameters are most sensitive to centered versus off centered data?

Which hyper-parameters of a convolutional neural network are likely to be the most sensitive to depending on whether the training (and test and inference) data involves only accurately centered images ...
2
votes
1answer
44 views

How GoogleNet actually deal with reducing overfitting?

Today I was going through a tutorial of Andrew Ng about Inception network. He said that GoogLeNet's hidden layers are also good in prediction and it had somehow a regularization effect, so it reduces ...
0
votes
2answers
56 views

What is a front-end and back-end in the context of convolutional neural networks?

The title is mainly it really. I googled that phrase or variations of it and couldn't find a decent result.
0
votes
1answer
30 views

not sure if fine-tuned network is finely-tuned

I am practicing with Resnet50 fine tuning for binary classification task, here is my code snippet. ...
2
votes
0answers
32 views

How to understand my CNN's training results?

I created a multi-label classification CNN to classify chest X-ray images into zero or more possible lung diseases. I've been doing some configuration tests on it and analyzing its results and I'm ...
0
votes
0answers
19 views

Is it possible to combine multiple SVMs that were trained on sublayers of a CNN into one combined SVM?

I have created a CNN for use on the MNIST dataset for now (so I have 10 classes). I have trained SVMs on the sublayers of this trained CNN and wish to combine them into a combined SVM as to give a ...
1
vote
0answers
19 views

Hinton's Capsule network 16 dimensions

As you may know, Hinton's Capsule Network has been around for about 2 years now. https://arxiv.org/abs/1710.09829 Much ado has been made about how the Capsules output a vector (magnitude = ...
2
votes
1answer
32 views

Accuracy scores in a Deep Learning project

I'm using three pre-trained deep learning models to detect vehicles and count from an image data set. The vehicles belong to one of these classes ['car', 'truck', 'motorcycle', 'bus']. So, for a ...
5
votes
4answers
5k views

Traffic signs dataset

I'm looking for annotated dataset of traffic signs. I was able to find Belgium, German and many more traffic signs datasets. The only problem is these datasets contain only cropped images, like this: ...
0
votes
1answer
24 views

Problem extracting features from convolutional layer where the dimensions are big for feature maps

I have trained a convolutional neural network on images to detect emotions. Now I need to use the same network to extract features from the images and use the features to train an LSTM. The problem is:...
0
votes
1answer
19 views

Multicamera Tracking vs Single Fisheye Camera

Suppose you want to detect objects and also track objects and people. Is it better to train a model using a single fisheye camera or using multiple cameras that mimic the view of the fisheye camera? ...
1
vote
1answer
52 views

How many parameters are being optimised over in a simple CNN?

Okay so here's my CNN (simple example from a tutorial) along with some arithmetic to get the total number of free parameters. We've got a dataset of 28*28 grayscale image (MNIST). First layer is a ...
1
vote
1answer
71 views

Is there a simple way of classifying images of size differing from the input of existing image classifiers?

Most image classifiers like Inception-v3 accept images of about size 299 x 299 x 3 as input. In this particular case, I cannot resize the image and lose resolution. Is there an easy solution of ...
2
votes
1answer
516 views

Neural Network for Optical Mark Recognition?

I've created a neural net using the ConvNetSharp library which has 3 fully connected hidden layers. The first having 35 neurons and the other two having 25 neurons each, each layer with a ReLU layer ...
1
vote
1answer
31 views

How to explain peak in training history of a convolutional neural network?

I am training a simple convolutional neural network to recognize two types of 1024-point frequency spectra (FFT). This is the model I'm using: ...
1
vote
0answers
16 views

Flattened vector observation or convolutional neural network input?

This is more of a general question of how to model/preprocess 'visual' state-observations to an Agent in Reinforcement Learning that I'll illustrate with an example. Say you have a reinforcement ...
5
votes
1answer
199 views

Other Deep Learning Networks for Visual Place Recognition?

I am doing a project on Visual Place Recognition in Changing Environments. The CNN used here is mostly AlexNet, and a feature vector is constructed from Layer 3. Does anyone know of similar work ...
0
votes
2answers
41 views

How can I convert the probability score between 0 to 1 to another format?

I have trained a multi-class CNN model using fastai. The model splits out probabilites for each of the three classes, which, of course, sum up to 1. The class with highest probability becomes the ...
1
vote
1answer
23 views

How can max pooling be applied to find features in words?

I'm reading about max pooling in a dynamic CNN paper and I can see how it can help find features in images cause the pixel with the highest density gets pooled, but how does it help find features in ...
0
votes
0answers
14 views

Optimizer effects on neural network with two outputs

I'm confused about the following issue. Let assume that we have a neural network that takes one input and two outputs. I try to visualize my model like as follows: ...
1
vote
0answers
24 views

Imposing contraints on sequence of image classifications

Are there example implementations of networks that apply constraints across sequences of image classifications where class labels are ordinal numbers? For example, to cause the output of a CNN to ...
0
votes
0answers
13 views

Affine Transformations and Data Augmentation

If you have a very distorted video/image, would affine transformations of the images make object detection algorithms make more mistakes compared to a normal camera?