Artificial Intelligence

Questions tagged [convolutional-neural-networks]

Ask Question

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

663 questions
Newest
Active
Bountied
1
Unanswered
Filter by
Sorted by
Tagged with
0
votes
0answers
7 views

Advice for a beginner with mathematics/electrical engineering background

I don't really know if somebody else already asked something related or if it is a duplicate. I'm an engineering student with a general background (French École + Regular EECS Brazillian University ...
3
votes
1answer
50 views

What does “off-the-shelf” mean?

I encountered the phrase/concept off-the-shelf CNN in this paper in which authors used off-the-shelf CNN representation, OverFeat, with simple classifiers to ...
0
votes
0answers
16 views

How to improve recognition of distanced objects?

I am developing a model of object detection based on fast-rcnn architecture (transfer learning) in tensorflow object detection API. My problem is that created model happens to produce very good ...
0
votes
0answers
16 views

how to extract only the color feature from image through cnn? [closed]

I wanted to extract only the color feature from the image dataset. and i want to extract this feature through. can anybody tell how can i do this
0
votes
1answer
13 views

CNN High Variance across multiple trained models, what does it mean?

Background: I have a 2D CNN model that I am applying to a regression task with some uniquely extracted spectrograms. The specifics of the data set are mostly irrelevant and very domain specific so I ...
0
votes
1answer
46 views

How to handle extremely 'long' images?

After transforming timeseries into an image format, I get a width-height ratio of ~135. Typical image CNN applications involve either square or reasonably-rectangular proportions - whereas mine look ...
1
vote
0answers
21 views

NN for card game evaluation function

I've written an Monte Carlo Tree Search player for the game of Castle (AKA Shithead, Shed, Palace...). I have set this MCTS player to play against a basic rule based AI for ~30000 games and collected ~...
4
votes
0answers
38 views
+50

How to implement taylor expansion of progressive Convolutional NN?

I intend to implement CNN like progressively expanded neural network in Keras. The basic idea is the first input node can be decomposed into multiple nodes with different orders and coefficient, then ...
0
votes
0answers
30 views

What is the difference between using dense layers as opposed to convolutional layers in my networks when dealing with images?

I am thinking about developing a GAN. What is the difference between using dense layers as opposed to convolutional layers in my networks when dealing with images?
1
vote
1answer
41 views

Alternatives to U-Net for biomedical image segmentation

Soon I will be working on biomedical image segmentation (microscopy images). There will be a small amount of data (a few dozens at best). My question is really straightforward: is there a neural ...
2
votes
0answers
16 views

What is the difference between training a model with RGB images and using only the color channels separately?

What is the difference between training a model with RGB images and using only the color channels separately (like only the red channel, green channel, etc.)? Would the model also learn patterns ...
2
votes
0answers
13 views

Adding a dense layer after a conv2d layer in a convolutional autoencoder

I am trying to implement a convolutional autoencoder with a dense layer at the bottleneck do to some dimensional reduction. I have seen two approaches for this which arent particularly scalable. The ...
1
vote
1answer
16 views

How are the weights retained for filters for a particular class in a CNN?

I am new to CNN. What I have learned so far about the filters is that when we are giving a training example to our model, our model updates the weights by gradient descent to minimize the loss ...
5
votes
2answers
92 views

Aren't all discrete convolutions (not just 2D) linear transforms?

The image above, a screenshot from this article, describes discrete 2D convolutions as linear transforms. The idea used, as far as I understand, is to represent the 2 dimensional $n$x$n$ input grid as ...
0
votes
0answers
14 views

Interpreting I/O Transformation Matrix in Convolution

I've been reading this article on convolutional neural networks (I'm a beginner) - and I'm stuck at a point. What I understand: We have a 4x4 input, and want to transform it to a 2x2 grid. I'm ...
1
vote
0answers
23 views

Why do DeconvNet use ReLU in the backward pass?

Why does DeconvNet (Zeiler, 2014) use ReLU in the backward pass (after unpooling)? Are not the feature maps values already positive due to the ReLU in the forward pass? So, why do the authors apply ...
1
vote
1answer
20 views

Object detection: combine many classes into one?

I am trying to train a model that detects logos in documents. Since I am not really interested in what kind of logo there is, but simply if there is a logo, does it make sense to combine all logos ...
1
vote
0answers
13 views

How to output a filter of equal size to the original image in Fully Convolutional Neural networks

I'm trying to perform a segmentation task on images of multiple sizes using fully convolutional neural networks. Currently, I'm using efficientnet as a feature extractor, and adding a deconvolution/...
1
vote
0answers
18 views

What are the main points of the top-down vs bottom-up paradigm in neural networks?

I've been reading some papers on human pose estimation and I'm starting to see the terms top-down and bottom-up crop up a lot. For example in this paper: Our hourglass module differs from these ...
3
votes
1answer
32 views

Can neurons in MLP and filters in CNN be compared?

I know they are not the same in working, but an input layer sends the input to x neurons with a set of weights, based of these weights and the activation layer, it produces an output that can be fed ...
1
vote
0answers
22 views

Human Aggression Detection Community, Competition and dataset

I'm looking for a community or competition website related to human aggression detection using Deep Learning in a video. Also, I'm looking for a dataset of human aggression activities. Any ...
2
votes
0answers
27 views

How to predict an event (or action) based on a window of time-series measurements?

I have an input vector $X$, which contains a series of measurements within a period, e.g. 100 measurements in 1 sec. The goal is to predict an event, let's say, moving forward, backward or static. I ...
2
votes
1answer
41 views

If the point of the ResNet skip connection is to let the main path learn the residual relative to identity, why are there convolutional skips?

In the original ResNet paper they talk about using plain identity skip connections when the input and output of a block have the same dimensions. When the input and output have different dimensions ...
0
votes
0answers
13 views

Training single network for one-shot identification

I am trying to build a face recognition application and I have seen implementations such as dlib. I would like to build a siamese net, and my doubts are about the architecture. Since I am supposed to ...
1
vote
0answers
25 views

Why does model complexity increase my validation score by a lot?

I learned that when creating neural networks the go to was to overfit and then to regularize. However I am now in a situation where, when I make the model more complex (more layers, more filters, ...) ...
0
votes
1answer
42 views

If vanishing gradients are NOT the problem that ResNets solve, then what is the explanation behind ResNet success?

I often see blog posts or questions on here starting with the premise that ResNets solve the vanishing gradient problem. The original 2015 paper contains the following passage in section 4.1: We ...
2
votes
1answer
40 views

Do deeper residual networks perform better or worse?

If you have an $18$ layer residual network versus and a $32$ layer residual network, why would the former do better at object detection than the latter, if you have both models are training using the ...
3
votes
2answers
165 views

How to calculate the number of parameters of a convolutional layer?

I was recently asked at an interview to calculate the number of parameters for a convolutional layer. I am deeply ashamed to admit I didn't know how to do that, even though I've been working and using ...
0
votes
0answers
25 views

Image data generator for my (x, 12, 370, 235, 3) dataset [migrated]

I have made a model with 12 2D CNN inputs. I suddenly realized that the ImageDataGenerator from from keras.preprocessing import image don't accept inputs with ...
2
votes
0answers
24 views

Is there an efficient way of determining the layers with the best performance as feature extractors in GoogleNet?

I am using a caffe model of pre-trained GoogleNet trained on ImageNet from here for image retrieval task (place recognition, more specifically). I would like to know the layer with best performance ...
1
vote
0answers
16 views

Can denoising auto-encoders be convolutional and fully connected?

I have been reading lately on autoencoders a lot. I just wanted to summarize my understanding of denoising autoencoders. As far as I understand they can be Fully connected (in which case, they will ...
0
votes
2answers
89 views

Understanding CNN in a few sentences

I don't know if this is the right place to ask this question. If it is not, please tell me and I remove it. I've just started to learn CNN and I'm trying to understand what they do and how they do it....
1
vote
2answers
46 views

Validation loss is lower than Training loss

I am training a classifier to identify 24 hand signs of American Sign Language. I created a custom dataset by recording videos in different backgrounds for each of the signs and later converted the ...
0
votes
0answers
32 views

How to Visualise a CNN model using Python [migrated]

I am new to Deep Learning and have been trying to show a plot of the CNN architecture using Python in Google Colab. Besides importing the necessary libraries, I have noticed from other resource that ...
2
votes
0answers
39 views

Is there a way to add “focus” on parts of the image when using CNNs?

I'm building a CNN/3DCNN model that classifies hand gestures. The problem is that the actual gesture occupies only like 1% of the whole image. That means that an enormous amount of convolutional ...
1
vote
1answer
49 views

How can I merge outputs of two separate layers so that the overall performance improves?

I am training a combined model (fine-tuned VGG16 for images and shallow FCN for numerical data) to do a binary classification. However, the overall AUC score is not what I expected it to be. Image-...
1
vote
0answers
16 views

Dealing with very similar object classes in object detection

I'm working on an object detection problem using Faster R-CNN. I need to identify two object classes, and they are very similar to one another. Furthermore they are similar to a third type of object ...
2
votes
1answer
22 views

Is it a sign of overfitting when validation_loss dips and then goes up with increasingly bigger swings?

I am experimenting with a ConvNet to categorize images taken with a depth camera. So far I have 4 sets of 15 images each. So 4 labels. The original images are 680x880 16-bit grayscale. They are scaled ...
0
votes
0answers
27 views

How can I use a Keras trained model saved in a HDF5 file to make predictions? [migrated]

I recently got started with neural networks. I built a handwritten character prediction model using the extended MNIST dataset, sklearn, Keras, numpy and pandas. The main goal is to take and/or upload ...
1
vote
0answers
19 views

Face recognition model loss not decreasing

I wrote a script to do train a Siamese Network style model for face recognition on LFW dataset but the training loss doesnt decrease at all. Probably there's a bug in my implementation. Could you ...
1
vote
0answers
10 views

Convolutional Feature Encoding Methods in DCNN

In Computer Vision, feature encoding methods are used on pre-trained DCNN to increase the feature robustness to certain conditions such as viewpoint/appearance variations ref. I was just wondering if ...
4
votes
2answers
65 views

Is there any difference between the convolution operation applied to images and applied to other numerical 2D data?

Is there any difference between the convolution operation applied to images and applied to other numerical 2D data? For example, we have a pretty good CNN model trained on a number of $64 \times 64$ ...
0
votes
0answers
15 views

In Fast R-CNN how are input RoIs mapped to the respective RoIs in the feature map before RoI pooling?

I've been reading the Fast R-CNN paper My understanding is that the input to one forward pass is the whole input image plus a list of RoIs (generated by selective search or another region proposal ...
1
vote
1answer
34 views

Using three image datasets with different image sizes to train a CNN

I've just started with AI and CNN networks. I have two NIFTI images dataset, one with (240, 240) dimensions and the other one with (256, 132). Each dataset is front a different hospital and machine. ...
1
vote
0answers
22 views

How to adapt MTCNN to large images with relatively small ROIs

This question could be generalised to how to adapt state-of-the-art object detection models to large images with small ROIs. In my particular case I'm trying to use this implementation of MTCNN to ...
1
vote
1answer
16 views

How do 3 channels affect a network when detecting human skin (CNN)?

Yeah I know, best title ever. Anyway, I want to make a neural network which is fed with frames coming from an usb camera. Don't wanna be so specific, so I'm just gonna say that the network's goal is ...
1
vote
1answer
27 views

Is it legal to construct a public image database (for deep learning) with images from the internet? [closed]

I am trying to put together a public agricultural image database of corn and soybeans, to train convolutional neural networks. The main method of image collection will be through taking pictures of ...
1
vote
0answers
118 views

What's the difference in using multiple convolutional layers and no pooling versus using a single convolutional layer and a single max pooling layer?

I'm currently working on a college project in which I'm designing a Deep Q-Network that takes images/frames as an input. I've been searching online to see how other people have designed their ...
2
votes
2answers
58 views

Should I apply image processing techniques to the inputs of convolution networks?

After working for some time with feature-based pattern recognition, I am switching to CNN to see if I can get a higher recognition rate. In my feature-based algorithm, I do some image processing on ...

1
2 3 4 5
…
14