Questions tagged [convolutional-neural-networks]

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

391 questions with no upvoted or accepted answers
Filter by
Sorted by
Tagged with
2
votes
0answers
48 views

Object size identification and maximum number of classes with convolutional neural networks

I am working on a project that involves using a ConvNet to identify screws. I am able to train from scratch a ConvNet based on the first version of the inception network, but shallower (only 3 ...
2
votes
0answers
12 views

How to debug and find neurons that most influenced a pixel in the output image?

I'm building CNN network of Image to Image. After training, I have some bad results in part of the Image. I would like to find the neurons that most influenced those pixels and do retraining only ...
2
votes
4answers
210 views

Can bounding boxes further improve the performance of a CNN classifier?

Suppose I have a standard image classification problem (i.e. CNN is shown a single image and predicts a single classification for it). If I were to use bounding boxes to surround the target image (i.e....
2
votes
0answers
130 views

What are the key differences between cellular neural network and convolutional neural network?

What are the key differences between cellular neural networks and convolutional neural networks in terms of working principle, implementation, potential performance, and applicability?
2
votes
0answers
43 views

Algorithms to indentify people in pictures without using face recognition

There are lot of researches about face detection in pictures, but is it the only way one can say "this person I'm looking for is here in this picture"? Aren't there algorithms that you can provide ...
2
votes
0answers
202 views

Convolutional Sequence to Sequence Learning: Training vs Generation

I am struggling to understand the use of the Convolutional Sequence to Sequence (Conv-Seq2Seq) model. The image below is take directly from the paper and is the nearly canonical diagram of the ...
2
votes
0answers
31 views

Is making lot of 1 versus other model efficient?

I've got classification problem on image, I have 10 classes and when I fine tuned my model on it (I tried VGG, Xception, resnet etc) I have approximatly 83% validation accuracy. I was wondering if ...
2
votes
0answers
30 views

Calculating tangent vector of curve s(P,$\alpha$) at given point $\alpha$ = 0

I am reading the paper "Transformation Invariance in Pattern Recognition – Tangent Distance and Tangent Propagation", where the tangent vector is calculated for the given curve $s(P,\alpha)$ at $\...
2
votes
2answers
109 views

Which neuron represents which part of the input?

In a neural network, each neuron represents some part of the input. For example, in the case of a MNIST digit, consider the stem of the number 9. Each neuron in the NN represents some part of this ...
2
votes
0answers
17 views

Sample from a distribution inside a NN layer

Is it possible to sample from a distribution inside a neural network forward function? Assume that there is a NN and a sample is needed to be derived from it at every forward-pass to randomly set a ...
2
votes
0answers
152 views

How can VAE have near perfect reconstruction but still output junk when using random noise input

I am creating a VAE for time series data using CNNs. The data has 4800 timesteps and 4 features. It is standardized and normalized. The network I am using is implemented in Keras as follows. I have ...
2
votes
0answers
38 views

How does a neural network output text box location data?

I'm interested in creating a convolutional neural network or LSTM to locate text in an image. I don't want to OCR the text yet, just find the text regions. Yes, I know Tesseract and other systems can ...
2
votes
0answers
45 views

How a game playing agent could identify potential objects and proximity?

Most implementations I'm seeing for playing games like Atari (usually similar to DeepMind's work using DQN) have 4 graphical frames of input fed into 3 convolutional layers which are then fed into a ...
2
votes
0answers
52 views

Using features extracted from a CNN as convolutional filter

I'm a bit confused about this. Assume I have a CNN network with two branches: Top Bottom The top branch outputs a feature vector of shape 1x1x1x10 (batch, h, w, c) The bottom branch outputs a ...
2
votes
0answers
37 views

Autoencoder why it is special for image decoding?

I have read about auto encoder. Understood what is encoding part, and decoding part, and the latent space. Now, i tried to implement this in keras. Below is the code. ...
2
votes
0answers
12 views

How do we stack two U-Nets to yield one final prediction?

I am trying to reproduce the model described in the paper DocUNet: Document Image Unwarping via A Stacked U-Net, i.e. stacking two U-Nets to yield one final prediction. The paper mentions that: ...
2
votes
0answers
31 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 ...
2
votes
0answers
81 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 ...
2
votes
0answers
179 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
0answers
40 views

The relationship between CNN terms

I'm new to CNNs and am wondering if I understand the relationship between the following terms: In image analysis, receptive fields group "input neurons" to reduce the connection to the next layer. ...
2
votes
0answers
136 views

Questions regarding keras activation maximization visualization

I wanted to use the visualization of the activation maximization of the filters that is described in the following keras tutorial/blog: https://blog.keras.io/how-convolutional-neural-networks-see-the-...
2
votes
0answers
36 views

FIlling space with empty bounding box

I'm detecting objects on images. I want to detect up to 10 objects, however, I'm not sure how to deal with the situation, where only one object is present. Should I fill the remaining spaces in the ...
2
votes
0answers
239 views

Deep NN architecture for predicting a matrix from two matrices

Recently my friend asked me a question: having two input matrices X and Y (each size NxD) where D >> N, and ground truth matrix Z of size DxD, what deep architecture shall I use to learn a deep model ...
2
votes
0answers
46 views

Regarding Tensorflow: How to Avoid Duplicate Use of Scope/Variable_names

I am trying to train Chess data through CNN. To proceed reinforcement learning, I had divided into two - "current network" and "reinforcement network". For each checkpoint file stored in different ...
2
votes
0answers
354 views

Similarity of images (CBIR) with CNN features

I am trying to build a neural network suitable to measure similarity between pairs of images. In particular I am interested in shoes. I have a query image (e.g. a shoe that I just took a picture of) ...
2
votes
0answers
47 views

Recommendations on which architecture to use to guess appointment

I'm currently developping an application which allows psychologists to manage their schedule and budget. As a proof of concept, I would like to create an intelligent appointment service. There can be ...
2
votes
0answers
208 views

Can anybody explain such behavior of accuracy and loss of my Net(caffe)?

I used this project for example(framework - caffe, arhitecture of net - mod of AlexNet, 400 images are used for training). I have this result: or this: Solver: ...
2
votes
0answers
387 views

CNN attention maps on non-images

My datasets are not actual images, so using methods with ImageDataGenerator or pre-trained networks might not apply in this case. Data Structure: Each "image" is a 2048-long vector that has float ...
2
votes
0answers
550 views

3D - CNN. Why my cost function decreases, but the accuracy does not increase?

I'm implementing a C3D-inspired neural network for human emotion recognition, the problem I'm facing is that altough the cost function is decreasing, for both training and validation sets, I do not ...
2
votes
1answer
117 views

Why do the inputs and outputs of a convolutional layer usually have the same depth?

Here's the famous VGG-16 model. Do the inputs and outputs of a convolutional layer, before pooling, usually have the same depth? What's the reason for that? Is there a theory or paper trying to ...
1
vote
0answers
32 views

Understanding Batch Normalization for CNNs

I am trying to understand how batch normalization (BN) works in CNNs. Suppose I have a feature map tensor $T$ of shape $(N, C, H, W)$ where $N$ is the mini-batch size, $C$ is the number of channels, ...
1
vote
1answer
46 views

How to extract parameters from a text using AI/NLP

lets say I have three texts: "make a heading that says hello word" "make a heading of hello world" "create heading consist of hello world" How can I fetch those groups ...
1
vote
0answers
20 views

In the DeepView paper, do they use the same FCN for all depth slices AND all views?

I'm trying to replicate a paper from Google on view synthesis/lightfields from 2019: DeepView: View Synthesis with Learned Gradient Descent and this is the PDF. Basically the input to the neural ...
1
vote
0answers
18 views

Is there any research work that shows that we should explicitly mark the word boundaries for 1D CNNs?

I'm doing character embedding for NLP tasks using one-dimensional convolutional neural networks (see Chiu and Nichols (2016) for the motivation). I haven't found any empirical evidence of whether or ...
1
vote
0answers
57 views

Are Graph Neural Networks generalizations of Convolutional Neural Networks?

In lecture 4 of this course, the instructor argues that GNNs are generalizations of CNNs, and that one can recover CNNs from GNNs. He presents the following diagram (on the right) and mentions that it ...
1
vote
1answer
53 views

Can someone explain me what does this loss curve says?

I was training a CNN model on TensorFlow. After a while I came back and saw this loss curve: The green curve is training loss and the gray one is validation loss. I know that before epoch 394 the ...
1
vote
0answers
21 views

How do I infer exploding or vanishing gradients in Keras?

It may already be obvious that I am just a practitioner and just a beginner to Deep Learning. I am still figuring out lots of "WHY"s and "HOW"s of DL. So, for example, if I train a ...
1
vote
0answers
28 views

What are the benefits of using ELU over other activation functions in CNNs?

I have come up with some examples of CNNs (segmentation CNNs) that use ELU (exponential linear unit) as an activation function. What are the benefits of this activation function over others, such as ...
1
vote
0answers
22 views

How can I improve the performance on unseen data for semantic segmentation using an auto-encoder?

I am using simple autoencoders for the task of semantic segmentation on the VOC2012 dataset. I am currently using a simple autoencoder based model. It is trained on adam optimizer with cross-entropy ...
1
vote
0answers
25 views

Is the 3d convolution associative given that it can be represented as matrix multiplication?

I'm trying to understand if a 3D convolution of the sort performed in a convolutional layer of a CNN is associative. Specifically, is the following true: $$ X \otimes(W \cdot Q)=(X \otimes W) \cdot Q, ...
1
vote
0answers
28 views

What is a “center loss”?

I have seen that a center loss is beneficial in computer vision, especially in face recognition. I have tried to understand this concept from the following material A Discriminative Feature Learning ...
1
vote
1answer
50 views

Single-Shot Learning for Object Re-Identification

I am looking for a way to re-identify/classify/recognize x real life objects (x < 50) with a camera. Each object should be presented to the AI only once for learning and there's always only one of ...
1
vote
0answers
38 views

How to have closer validation loss and training loss in training a CNN

I am using an AlexNet architecture as my Convolutional Neural Network. A learning rate of 0.00007 and 128 batch_size. I have 20000 data and 10% test, 40% validation, and 50% for training. I used 100 ...
1
vote
0answers
51 views

Computer vision - Can you put more weight on a specific part of the object?

Let's say I'm looking for any item that has a certain shape (outline) in a photo. but I can further classify it only according to particular features, that most of them are expected to be shown only ...
1
vote
0answers
34 views

Training a CNN for semantic segmentation of large 4600x4600px images

I am trying to implement a CNN (U-Net) for semantic segmentation of similar large grayscale ~4600x4600px medical images. The area I want to segment is the empty space (gap) between a round object in ...
1
vote
0answers
18 views

Model output segmentation maps which are not full

I created a VGG based U-Net in order to perform image segmentation task on yeast cells images obtained by a microscope. There are a couple of problems with the data: There is inhomogeneity in the ...
1
vote
1answer
40 views

Pytorch Deep q network not learning and step not stepping towards target

I am trying to create a simple deep q network for rl with conv2d layers. I can’t figure out what I am doing wrong, and the only thing I can see that doesn’t seem right is when I get the model ...
1
vote
0answers
20 views

Duplicating calculations in CNN-LSTM architecture

I want to use frames from video game and analyze them using CNN and LSTM. But when I have the model defined like that ...
1
vote
1answer
37 views

How to construct input dependent convolutional filter?

I am constructing a convolutional variational autoencoder for images, starting out with mnist digits. Typically I would specify convolutional layers in the following way: ...
1
vote
1answer
106 views

Are mult-adds and FLOPs equivalent?

I am comparing different CNN architectures for edge implementation. Some papers describing architectures refer to mult-adds, like the MobileNet V1 paper, where it is claimed that this net has 569M ...

1 2
3
4 5
8