Skip to main content

Questions tagged [cnn]

Convolutional Neural Networks (CNN, also called ConvNets) are a tool used for classification tasks and image recognition. The name giving first step is the extraction of features from the input data.

481 questions with no upvoted or accepted answers
Filter by
Sorted by
Tagged with
4 votes
0 answers
271 views

ResNet50 + Transformer

In many papers people extract features from image using ResNet and than pass them through transformer. I want to implement the same. I want to get features and than classify them using transformer. ...
alex-uarent-alex's user avatar
4 votes
1 answer
315 views

why transpose convolution is called "transpose"

https://d2l.ai/chapter_computer-vision/transposed-conv.html https://en.wikipedia.org/wiki/Transpose I understand what transpose convolution does, but I am confused about the name of 'transpose'. In ...
richard.g's user avatar
  • 141
4 votes
2 answers
1k views

How does ResNet bottleneck architecture's input size is possible to change from 56x56x64 to 56x56x356?

In ResNet papaer, First residual block's input size is 56x56x64 caused by 7x7x64 filter in first layer. But, in the paper, they showed residual block that has 56x56x256 input size. How does it is ...
douner's user avatar
  • 125
4 votes
1 answer
97 views

Training Machine Learning Model - Neural Network - Islands Problem

I was working on the following leetcode problem: Given a 2d grid map of '1's (land) and '0's (water), count the number of islands. An island is surrounded by water and is formed by connecting ...
user87771's user avatar
4 votes
1 answer
312 views

How does Pooling Layer in CNN introduce invariance to other transformations besides translation

Here is a quote from deeplearningbook which I am trying to process. I am not sure what do they mean by this quote, can someone help me understand please? Pooling over spatial regions produces ...
Stefan Radonjic's user avatar
4 votes
3 answers
951 views

How to determine the number of the training images in Keras after data augmentaion?

I want to create a CNN model and I am using data augmentation. I want know the number of augmented images in Keras. How to determine the number of the training images in Keras after data augmentation?...
N.IT's user avatar
  • 1,995
3 votes
2 answers
2k views

Keypoint detection from an image using a neural network

I am trying to design and train a neural network, which would be able to give me coordinates of certain key points in the image. Dataset I've got a dataset containing 1800 images similar to these: ...
Ladislav Ondris's user avatar
3 votes
0 answers
240 views

How to create anchor-positive and anchor-negative pair for feature X in a signature data set for training Siamese network

How to create anchor-positive and anchor-negative pair for feature X in a signature data set for training Siamese network? Im have a cedar signature data set with 55 peoples signatures(classes) with ...
star's user avatar
  • 1,481
3 votes
0 answers
26 views

Why RANDOM noise images always predicted as BIRD?

Say I have fine-tuned a 10-classification ResNet18 network on CIFAR-10 and the accuracy on validation set is about 93%. However when feeding into 5000 random noise images (Gaussian noise with the ...
dmrak's user avatar
  • 31
3 votes
0 answers
118 views

Computing derivatives for backpropagation across a convolution step

This will be a long post, but I hope it'll be instructive to anyone else in my position. I'm trying to find how the derivatives of the loss function are calculated with respect to the kernels and ...
Shirish's user avatar
  • 299
3 votes
0 answers
950 views

Bounding box regression in R-CNN

In R-CNN paper, they give the definition of the target values for bounding box regression Given that $(P, G)$ is a (prediction box, ground-truth box) pair of the form $(x, y, w, h)$ where $x, y$ is ...
HOANG GIANG's user avatar
3 votes
0 answers
337 views

convLSTM : how to structure input data

I have the following dataframe containing training data that I have been using to perform a regression task using CNN + FC : ...
FenryrMKIII's user avatar
3 votes
2 answers
760 views

LeNet-5 - combining feature maps in C3 layer

Famous LeNet-5 architecture looks like this: The output of layer S2 has dimension: 10x10x6 - so basically an image with 6 convultions applied to it to derive features. If each dimension was again ...
Mateusz Konopelski's user avatar
3 votes
2 answers
215 views

How to fetch text from pdf to further proceed with question answer based model from the same document?

To illustrate the above title. Suppose you have a pdf document, which is basically scanned from hardcopy, now there are set of fixed questions to answer from the document itself. For an example a ...
Arijit Das's user avatar
3 votes
0 answers
206 views

Identifying computer scanned digits

I have digit images as below which I would like to identify: Some are of slightly worse quality : The images are not of a fixed resolution but are mostly in the range (80*20 to 130 *40). Due to ...
Syed Saad's user avatar
3 votes
0 answers
49 views

Formula to calculate size of Capsule output similar to the formula for CNN?

Is there any formula to find the output dimensions of a capsule network similar to that of a Convolutional Neural Network? For Example: In CNN, we know that ...
Reuben_v1's user avatar
  • 151
3 votes
0 answers
519 views

Deep CNN with variable number of classes and "vanishing" data

I am using a deep CNN to predict the class an image belongs to (N classes). However, the number of classes is not stationary. I.e. over the time the network will be used, some new classes may emerge ...
sist's user avatar
  • 131
2 votes
1 answer
723 views

Input shape of dataset in hybrid CNN-SVM classifier

I am working on hybrid CNN-SVM for classification task, where I aim to use CNN for feature extraction and SVM for classification. So after the training of my CNN model as below: ...
root's user avatar
  • 135
2 votes
0 answers
70 views

Which model is used for document extraction (CamScanner, Microsoft Lens etc)

I want to start a small project where I'd create a model(s) that would extract document from a picture and rescale it, something like CamScanner or Microsoft Lens apps do. I've gathered a small ...
apantovic's user avatar
2 votes
1 answer
313 views

Is this Double U-Net overfitting?

I'm working on a undergraduate project with using deep learning. Currently, I'm trying to improve a model by modifying it. Model is Double U-Net and dataset that I'm using is DRIVE dataset. It ...
Mustafa Tufan's user avatar
2 votes
1 answer
2k views

To freeze or not, batch normalisation in ResNet when transfer learning

I'm using a ResNet50 model pretrained on ImageNet, to do transfer learning, fitting an image classification task. The easy way of doing this is simply freezing the conv layers (or really all layers ...
amateurjustin's user avatar
2 votes
0 answers
110 views

Can Darknet be Integrated with Tensorflow and tfx?

Tensorflow and Darknet are deep learning frameworks that work and are configured. Can Darknet be Integrated differently? Is there any framework or way to integrate Darknet within a tfx pipeline, for ...
pentanol's user avatar
  • 161
2 votes
0 answers
259 views

Applying GradCam to video classification models

In the original paper, it says that GradCam visualization can be applied to any convolution based model. The problem is stated for convolutions that process images. In my case, I am classifying videos ...
Iván Mindlin's user avatar
2 votes
0 answers
106 views

Methods to visualize the filters in the later layers of a CNN?

I've extracted the weights from the filters of a pretrained model (AlexNet). I wish to represent these weights visually, this works fine for the first layer as there is only 3 input channels so I can ...
Fraser Hamilton's user avatar
2 votes
1 answer
705 views

Deep learning test loss curve won't go down

I've been working with Deep Learning projects for this current project that I am working on and it's basically a time series classification problem. Where given an array of time series data I need to ...
Joseph Anderson's user avatar
2 votes
1 answer
214 views

feature importance with CNN

I can use shap to extract important features for Dense NN. However, for CNN, I encountered two problems: the feature order may be messed up or combined after the ...
user110020's user avatar
2 votes
0 answers
803 views

Train and valid accuracy and loss stay the same over many epochs of training with Pytorch

I am building a neural network to recognize hand gestures from the leapGestRecog data set from Kaggle. While training I ran into some issues. Here are some images of my data set: I augmented my data ...
Mariusz's user avatar
  • 21
2 votes
0 answers
43 views

out of frame landmark detection with CNN

a robust landmark detector must cope with occlusion (the landmark surrounding are occluded by another object) the detection can still be performed but it should be stressed that the landmark is ...
yuri's user avatar
  • 23
2 votes
1 answer
52 views

Dealing with high dimensionality datasets

I have data of dimensionality (25000, 100, 500) i.e. 25000 rows each consisting of a 2 dimensional 100 X 500 matrix. Currently I am only applying CNN for ...
user75252's user avatar
2 votes
0 answers
51 views

Help required in understanding how the error of a convolutional layer is calculated when filter and delta of next layer have differing dimensions

I am trying to implement a CNN in NumPy so as to better understand its inner workings My architecture is as follows 10 images with 1 channel and with 28-pixel rows and columns (Dimension : (...
adhok's user avatar
  • 121
2 votes
1 answer
4k views

Getting confusion matrix with Keras flow_from_directory

For a homework I have to analyse a set of images. For this I plan to use convolutional neural network. The images are split onto specific folders : A test set with 624 photos dataset/test/normal (...
davidvera's user avatar
  • 121
2 votes
0 answers
130 views

Understanding CNN by visualizing class activations using GRAD_CAM

I followed the blog Where CNN is looking? to understand and visualize the class activations in order to predict something. The given example works very well. I have developed a custom model using ...
Ali Raza Memon's user avatar
2 votes
0 answers
20 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 ...
Alexander Soare's user avatar
2 votes
0 answers
243 views

Keras: DepthwiseConv3DTranspose or doing transposed Conv. with a Conv. layer

I am building an autoencoder for 3D images and would like to use Depthwise convolutions. For the encoder, I found an implementation of a depthwise 3D convolutional layer (DepthwiseConv3D). For the ...
PascalIv's user avatar
  • 432
2 votes
1 answer
137 views

What doese 'v' mean in GoogLeNet?

In GoogLeNet (this link), there is 'v' notation in Figure3 like '1X1+1(v)'. I don't know the meaning of 'v'. Also, I understood 's' as stride. But, I don't know the reason why plus operation is used ...
douner's user avatar
  • 125
2 votes
0 answers
58 views

Size difference during backpropagation between fully connected layer and convolution layer?

This is a simple example of a network consisting of two convolutional layers and one fully connected layer. ...
Tion's user avatar
  • 21
2 votes
0 answers
57 views

Load test data to pass into predict_classes()

I am using the Sequential model to train a CNN. The test data are in the submission_data_path as .png images. Is it possible to ...
hdk's user avatar
  • 121
2 votes
0 answers
2k views

How to solve warning : ImageDataGenerator specifies `featurewise_center`, but it hasn't been fit on any training data?

I am doing image classification and I have a training set and a test set with different distributions. So to try to overcome this problem I am using an Image generator in the following way: ...
J.D.'s user avatar
  • 881
2 votes
0 answers
23 views

Different convolutions in CNN

I have a simple question. Why only convolution is used in CNN? There are a lot of possible rules for combining a filter and an image. Why is pixel-wise convolution the standard? For example, dropout ...
Charles Martin's user avatar
2 votes
3 answers
359 views

Can the same CNN architecture be used for different data sets?

I have a CNN architecture that works well on 32x32x3 images. Can I use that same architecture for a data set made up of 28x28x1 images? (Both data sets have 10 classes). If this is possible, what ...
brubrudsi's user avatar
2 votes
1 answer
45 views

How to find which patch in orignal image does an activation correspond to in vgg net after the final pooling layer

So I am working on the NeurIPS 2019 reproducibility challenge, The link to the paper is https://arxiv.org/abs/1806.10574. So basically we have a vgg-16 net with the final fully-connected layers ...
AYUSH MANGAL's user avatar
2 votes
0 answers
54 views

Multilabel Classification; which network design?

I have a hard time thinking about how I can build this network with the following problem: I want to build a CNN to classify notes from sheet music. I have tried several models with and without ...
Jordy's user avatar
  • 31
2 votes
0 answers
32 views

List of CNN for Emotion/Sentiment recognition on images with performance on main datasets (IAPS, GAPED, EmoPics, NAPS)

There are more and more databases of pictures classified or rated with emotions. For instance, I know of 4 databases (IAPS, GAPED, EmoPics, NAPS) rating pictures on 2 dimensions: Valance (positive vs ...
RemiDav's user avatar
  • 121
2 votes
0 answers
442 views

Why do UNET-like architectures perform better than sliding-window approaches?

I'm writing a thesis that heavily focuses on semantic segmentation of biomedical images. I'm reviewing different segmentation approaches, identifying two main approach branches: A sliding window-...
Filippo Castelli's user avatar
2 votes
2 answers
92 views

Does high accuracy metrics with small (but equally sampled) dataset means a good model?

I have been training my CNN with 200 images per class for a classification problem. There problem is a binary classification one. And with the amount of test data ( 25 per class) I am getting good ...
Sangathamilan Ravichandran's user avatar
2 votes
0 answers
2k views

Transfer learning on yolo using keras

I am working on a project that uses object detection. I have logo images that need to be detected in a video. I am doing this in keras. I followed this blog to convert the yolo weights to a keras ...
Ashu's user avatar
  • 43
2 votes
1 answer
920 views

How to put data into a 1-dimensional ConvLSTM2D with keras?

I am attempting to adapt the frame prediction model from the keras examples to work with a set of 1-d sensors. I have android wearable sensor data and am designing an algorithm that can hopefully ...
Chris Cabral's user avatar
2 votes
0 answers
43 views

Trained vs randomly initialised cnn model. Difference in inference speed

We have 2 models written in Keras. We then convert them to coreml format and test the inference speed of each one on iPhone 7. The model architecture is the same for each of the two. The 1st model is ...
Victor Khodalov's user avatar
2 votes
0 answers
249 views

Audio files and their corresponding spectrograms for image classification process

Suppose I have a dataset of audio files that I have to use for whale sound classification. I am choosing the strategy of treating it as an image classification problem by using their corresponding ...
Abhishek Singh's user avatar
2 votes
1 answer
112 views

Model for classifying time-series data with distinct features?

I've heard about time-series classification being done with TCN's and CNN's combined with LSTM's very often, citing that CNN's would provide insight both forward and in the past since you already have ...
Rithwik Sudharsan's user avatar

1
2 3 4 5
10