Questions tagged [computer-vision]

Questions related to image representation, segmentation, visual object categorization and image processing algorithms in general.

Filter by
Sorted by
Tagged with
1
vote
1answer
1k views

TensorFlow Object Detection - Multiple objects detection with the pet detector

I have trained the pet detector from tensorflow object detection with a ssd mobilenet architecture. The train dataset only contains image with a single objects (1 image = 1 box ). I would like to ...
1
vote
1answer
139 views

Satellite data pre-processing for Keras CNNs [closed]

I’m looking at satellite data and want to do object detection using CNNs in Keras. I’m currently pre-processing the data (turning them into tensors) that I’ve obtained which include the original ...
1
vote
0answers
105 views

What is the relationship between Eigenface (PCA) & Fisherface (FLDA) on $S_{x} \bigcup S_{y}$?

I have a question about the relationship between the result of Eigenface and Fisherface on face recognition. (Eigenface and Fisherface is the topic of Computer Vision. Those are related with PCA and ...
1
vote
0answers
25 views

Identify people by recognizing their entire bodies [closed]

we're working on project where people try to find photos of themselves from large albums. Inorder to make their job easy, we used AWS Rekognition for facial recognition and automatically suggesting ...
1
vote
0answers
466 views

Batch normalization: is the same gamma and beta used throughout a convolutional layer feature map?

In the batch normalization paper it is mentioned that the same gamma and beta are used throughout a feature map for convolutional layers. I interpret this as the filter weights are consistently used ...
1
vote
1answer
906 views

Best ratio of negative example to create for a multiclass problem

I am working on a multi-class classification problem using images. I have a training library of images containing 9 different classes of object, however I will also need to train my image classifier ...
1
vote
0answers
270 views

An algorithm to read handwriting from checks

I've noticed that ATM machines have become very good at reading handwriting on checks. I would like to write a program (using some appropriate machine learning or computer vision library) that is ...
1
vote
1answer
129 views

Training CNN - 3 different training sets, different noise

I am doing a road segmentation task for high resolution images. I have three different data sets: Around 100 with extra high resolution with the ground truth. Around 500 images with slightly worse ...
1
vote
1answer
897 views

SegNet CamVid dataset training classes mismatch?

This is with reference to the CamVid dataset and one of its tutorial here: http://mi.eng.cam.ac.uk/projects/segnet/tutorial.html I'm quite confused by how the model is supposed to be trained on 11 ...
1
vote
0answers
385 views

How to calculate the distance histogram between two specific super pixels?

I am very new to image processing and Matlab. I am working with RGB image and used SLIC algorithm to generate superpixel for an image. By using regionprops to calculate the superpixel properties. <...
1
vote
0answers
33 views

How are multi-class losses computed in a Deconvolution Network?

I am wondering what is the math and possible code implementation of computing a multi class loss in a Deconvolution Network based on this paper. Basically, if we have an image that is fed into the ...
1
vote
0answers
873 views

3d Convolution vs CNN-LSTM for Gesture recognition

I want to implement a gesture recognition system from video (of hand movements). Some people have experimented with 3d convolutions to extract not only spatial features out of images, but also ...
1
vote
1answer
394 views

Implementation of CNN [closed]

I am new to the field of vision. To get a good understanding of the concepts, I wanted to look at source code of some CNN. Can anybody suggest some place where I can find implementation of simple CNN ...
1
vote
0answers
52 views

Can a Homography be used when images are slightly different?

I have two pictures of a mole, but second picture is slightly different from the first because mole has changed its form a bit. Both images have different perspective, so I am willing to compute a ...
1
vote
0answers
240 views

stacked autoencoder decoder and deconvolutional network

The convolution-deconvolutional architecture can be used for image segmentation or re-construction. These types of tasks can also be done using the architecture of stacked autoencoder-decoder. But ...
1
vote
0answers
32 views

Thresholding from a set of 2D points

In real world situations, most often the estimated 2D positions from the videos are effected by a noise. I have a set of consecutive 2D positions for moving objects in 2D space. Part of this data (...
1
vote
0answers
182 views

What are the current most challenging MNIST-like tasks aiming to achieve the lowest error rate?

For example there is the MNIST database which is used to test artificial neural network (ANN), however it's not so challenging, because some hierarchical systems of convolutional neural networks ...
1
vote
0answers
14 views

What's the relationship between the sift features and the minutiae features?

As we know, in the field of fingerprint or palmprint recognition, both the sift features and the minutiae features can be used to construct the descriptor. However, what's the relationship between ...
1
vote
0answers
152 views

Why does my Linear SVM on SIFT Features converge, but not a non-Linear SVM or a Convolutional Neural Network?

I have two distinct texture classes of 1000 images each with varying amounts of blur and noise (but with a reasonable threshold on both so that even if both are at their highest levels the class of ...
1
vote
0answers
227 views

Bag of Features: why the distance between two histograms of the same image is different than 0?

I'm trying to implement a Content Based Image Retrieval application for small image datasets. I'm testing it just with 1 thousands images from Caltech1001. The approach that I'm using is the classic ...
1
vote
0answers
493 views

CNN training and overfitting

When testing the training of a CNN code with a small data set (approx 2560 images each for training and validation), what is over-fitting and how can it be mitigated? Arnold
1
vote
2answers
168 views

How to think about the architecture of the Convolutional Neural Network?

Recently, I've started to learn more about CNNs to use them in some computer vision tasks. At the moment, I have roughly good knowledge about different parts of a CNN such as layers, solvers, loss ...
1
vote
0answers
55 views

what are some of the success stories of applying optimization in computer vision?

I'm trying to find several examples that demonstrate the success of optimization methods in application to computer vision and robotics. In particular, I'm interested in problems I can formulate from ...
1
vote
0answers
197 views

Terrain classification for self-driving vehicle

So I just started in researching in terrain classification for self-driving vehicles. Basically I wish to distinguish grass, road, building, people and mud etc from the camera. I've experience with ...
1
vote
0answers
619 views

Is there an alternative to PCA that produces a unique representation?

PCA produces solutions that are not unique i.e. the resulting representation could be recreated using a different set of points. This is where my problem lies. I was wondering whether anyone is aware ...
1
vote
0answers
2k views

Mean Average Precision in Matlab with liblinear and vlfeat

I want to find the mean average precision (meanAP) from a classification problem. I am using liblinear for classification and I am trying to use vlfeat for the precision because it already includes a ...
1
vote
0answers
278 views

why pretraining for convolutional neural networks [closed]

Usually Back propagation NN has the problem of vanishing gradients. I found that Convolutional NN (CNN) some how get rid of this vanishing gradient problems (why?). Also in some papers some ...
1
vote
0answers
1k views

Machine Learning Process for detecting edges of overlapping objects with OpenCV

I'm quite new to machine learning and a bit unsure about the whole process and the interpretation of the results. The Task: I have images with some objects of somewhat the same color and shape which ...
0
votes
4answers
3k views

Object localization with CNN [closed]

I am interested in locating the center of a playing card on the surface of a table: I have written a script so that I can generate images like this, where the card is moved around and rotated. My ...
0
votes
1answer
57 views

batch size of stochastic gradient descent [duplicate]

I understand that stochastic gradient descent has a batch size of 1, but while reading inception v2 paper, I found this text in training methodology "We have trained our networks with stochastic ...
0
votes
1answer
487 views

What was the batch size used in the original GoogleNet paper (Inception)?

It does not seem to be explicitly mentioned in the methods of the paper (available at https://arxiv.org/pdf/1409.4842.pdf).
0
votes
1answer
5k views

Vehicle license plate recognition using Convolutional Neural Network trained with mnist data

I would like to construct a license plate recognition system using convolutional neural network (CNN). But I do not have appropriate dataset to train from. If I train my CNN on the MNIST handwritten ...
0
votes
1answer
359 views

What is the difference between simple histogram and histogram LBP?

What is the difference between a simple histogram and a histogram LBP (local binary pattern)? Can someone provide the intuition behind histogram LBP?
0
votes
1answer
132 views

Why does distorting images improve training on a neural network?

I cannot understand why distorting an image, e.g flipping it, increasing the gamma intensity would somehow increase the accuracy on neural network. Within my situation, I am Using a CNN to detect if ...
0
votes
1answer
253 views

How yolo9000 predicts big objects?

I have a theoretical question about Yolo9000. Since it's a fully convolutional neural network, I was wondering how can it predict large objects. To the best of my knowledge, each convolutional kernel "...
0
votes
1answer
323 views

In what form is optical flow used an input to a neural network? [closed]

I have extracted the optical flow of images. Should I pass the x,y optical flow or should I pass the RGB visualization as the input. What is the difference between the two?
0
votes
1answer
199 views

Understanding neural networks and classes

I want to know if this argumentation is valid or not of my algorithm. I'm trying to implement a CBIR (Content-Based Image Retrieval) where I've used the basics on CBIRs (colour, texture, shape, ...
0
votes
1answer
561 views

multi-class logistic regression for ordered labels

I have a set of labells: A - no lesion B - mild C - severe If an instance from a class predicted as its nearest class not a big problem ...
0
votes
2answers
37 views

How can I tell if images are from the same distribution?

What are some tests I could run on two groups of imagery to see if they're from the same distribution? Imagine, for example, that you want to augment a dataset with another, and you want to make sure ...
0
votes
1answer
22 views

Analysing the impact of each CNN layers

I want to analyse the impact of each layer of CNN. I have trained the CNN model with a dataset. After that, weights of first convolutional layer are fixed and remaining layers are initialise to zero ...
0
votes
1answer
34 views

Object classification

I'm currently working on a "Where's Waldo" project as part of my coursework, where I have to find 3 different characters in any given image - Waldo, Wenda, and Wizard. I'm trying to convert this ...
0
votes
1answer
66 views

How do convolutional neural networks learn from images of different translations and conditions?

when we feed the CNN images of cats in different lighting conditions or colors, Is it the job of the conv layers to learn the different representations(lighting conditions and colors) and map them to ...
0
votes
1answer
33 views

How to correctly find the best hyperparameter combination when training a neural network?

I am not sure whether this is the right place to ask this question, so feel free to redirect me if not. What I'm doing is bench-marking a model (MobileNet v2 100 224) in terms of performance - size ...
0
votes
1answer
100 views

How to deal with different scales with CNN?

In the deep learning book by Goodfellow et al., it is stated "Convolution is not naturally equivariant to some other transformations, such as changes in the scale or rotation of an image. Other ...
0
votes
1answer
232 views

When doing data augmentation, should you train with the original data as well or just the augmented data?

When doing data augmentation in computer vision problems, should you train with the original (un-augmented) data as well or just the augmented data? Are there pros and cons to the two strategies or ...
0
votes
1answer
38 views

How does smoothing an image gives it a different scale?

In some deep learning papers i read about multiscale inputs, so i wanted to read about scale of an image. What i got to know is that fundamentally scale is related to the distance of the object being ...
0
votes
1answer
326 views

Introduction to Conditional random fields

I came across the application of a conditional random field (CRF) to the output from a convolutional neural network (CNN) for image segmentation. The additional CRF step seems to be a common ...
0
votes
1answer
515 views

How does Batch Normalization not lead to the model blowing up? [duplicate]

I was reading the Batch Norm paper and in this paragraph, We could consider whitening activations at every training step or at some interval, either by modifying the network directly or by ...
0
votes
1answer
400 views

Unbalance images dataset

I want to create a deep learning model to classify images. My dataset has around 400 classes and the classes have different number of images.. How can I train the deep learning network on unbalanced ...
0
votes
1answer
118 views

how does the addition of 1×1 layers reduce from the complexity and solve sparse structure in inception module?

Based on the (going deeper with convolutions) paper. The inception module consists of multiple branches, each branch consists of convolution layer and each layer has a different kernel size for ...

1
3 4
5
6 7