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
43
votes
4answers
37k views

What is translation invariance in computer vision and convolutional neural network?

I don't have computer vision background, yet when I read some image processing and convolutional neural networks related articles and papers, I constantly face the term, ...
43
votes
7answers
7k views

Neural network references (textbooks, online courses) for beginners

I want to learn Neural Networks. I am a Computational Linguist. I know statistical machine learning approaches and can code in Python. I am looking to start with its concepts, and know one or two ...
17
votes
4answers
10k views

Is it possible to give variable sized images as input to a convolutional neural network?

Can we give images with variable size as input to a convolutional neural network for object detection? If possible, how can we do that? But if we try to crop the image, we will be loosing some ...
14
votes
3answers
13k views

hinge loss vs logistic loss advantages and disadvantages/limitations

Hinge loss can be defined using $\text{max}(0, 1-y_i\mathbf{w}^T\mathbf{x}_i)$ and the log loss can be defined as $\text{log}(1 + \exp(-y_i\mathbf{w}^T\mathbf{x}_i))$ I have the following questions: ...
14
votes
2answers
5k views

What is energy minimization in machine learning?

I was reading about optimization for an ill-posed problem in computer vision and came across the explanation below about optimization on Wikipedia. What I don't understand is, why do they call this ...
12
votes
1answer
11k views

How to form a Precision-Recall curve when I only have one value for P-R?

I have a data mining assignment where I make a content-based image retrieval system. I have 20 images of 5 animals. So in total 100 images. My system returns the 10 most relevant images to an input ...
12
votes
1answer
14k views

How to reduce number of false positives?

I'm trying to solve task called pedestrian detection and I train binary clasifer on two categories positives - people, negatives - background. I have dataset: number of positives= 3752 number of ...
11
votes
5answers
34k views

What loss function should I use for binary detection in face/non-face detection in CNN?

I want to use deep learning to train a face/non-face binary detection, what loss should I use, I think it is SigmoidCrossEntropyLoss or Hinge-loss. Is that right, but I also wonder should I use ...
11
votes
3answers
1k views

Convolutional Neural Network Scale Sensitivity

For the sake of example, lets suppose we're building an age estimator, based on the picture of a person. Below we have two people in suits, but the first one is clearly younger than the second one. (...
11
votes
1answer
3k views

Training a convolution neural network

I am currently working on a face recognition software that uses convolution neural networks to recognize faces. Based on my readings, I've gathered that a convolutional neural network has shared ...
10
votes
2answers
6k views

Difference between pooling and subsampling

At this point in a video on LeNet1, Yann LeCunn seems to make a distinction between pooling and subsampling, with a separate gesture for each: [...] The second version had a separate convolution ...
10
votes
3answers
9k views

How to classify a unbalanced dataset by Convolutional Neural Networks (CNN)?

I have a unbalanced dataset in a binary classification task, where the positives amount vs negatives amount is 0.3% vs 99.7%. The gap between positives and negatives are huge. When I train a CNN with ...
10
votes
2answers
5k views

Anchoring Faster RCNN

In the Faster RCNN paper when talking about anchoring, what do they mean by using "pyramids of reference boxes" and how is this done? Does this just mean that at each of the W*H*k anchor points a ...
10
votes
1answer
9k views

How to determine the number of convolutional operators in CNN?

In computer vision task, such as object classification, with Convolutional Neural Networks (CNN), the network provides an appealing performance. But I'm not sure how to set up the parameters in ...
10
votes
2answers
5k views

Balancing Reconstruction vs KL Loss Variational Autoencoder

I am training a conditional variational autoencoder on a dataset of faces. When I set my KLL Loss equal to my Reconstruction loss term, my autoencoder seems unable to produce varied samples. I always ...
10
votes
2answers
14k views

Can a convolutional neural network take as input images of different sizes?

I'm working on a convolution network for image recognition, and I was wondering if I could input images of different sizes (not hugely different though). On this project: https://github.com/...
9
votes
3answers
5k views

ImageNet: what does top-five error means?

One of the evaluation method for ImageNet Competition (classify 1,000 categories images) is top-5 error, what does that mean? See: http://www.image-net.org/challenges/LSVRC/
9
votes
1answer
5k views

Fine Tuning vs Joint Training vs Feature Extraction

I am reading this paper http://zli115.web.engr.illinois.edu/wp-content/uploads/2016/10/0479.pdf It distinguishes between feature extraction and fine tuning in deep learning. I am not getting the ...
9
votes
1answer
4k views

regarding the output format for semantic segmentation

While reading the semantic segmentation papers as well as their corresponding implementations, I found that some approaches use softmax while others use sigmoid for the pixel-level labeling. For ...
8
votes
2answers
8k views

Feature extracted by max pooling vs mean pooling

In deep learning, and it's application to computer vision, is it possible to tell what kind of features these two types of pooling extract? e.g. is it possible to say that max pool extracts edges? Can ...
8
votes
1answer
6k views

patch wise training and fully convolutional training in fully convolutional neural network

In the paper of fully convolutional neural network, the authors mention both patch wise training and fully convolutional training. My understanding for the training set construction is as follows: ...
8
votes
1answer
8k views

Deep Learning: Why does increase batch_size cause overfitting and how does one reduce it?

I used to train my model on my local machine, where the memory is only sufficient for 10 examples per batch. However, when I migrated my model to AWS and used a bigger GPU (Tesla K80), I could ...
7
votes
4answers
7k views

Open source classification algorithms, preferably in C++ [closed]

I am in search of open source classification algorithms. I am working on a computer vision project that uses classification for scene recognition. I wish to bench test a range of machine learning ...
7
votes
1answer
2k views

One-shot object detection with Deep Learning

In the recent years, the field of object detection has experienced a major breakthrough after the popularization of the Deep Learning paradigm. Approaches such as YOLO, SSD or FasterRCNN hold the ...
7
votes
1answer
2k views

Why do people use Zero-Padding in Convolutional Neural Networks?

I am wondering why people usually pad with zeros instead of e.g., using the min-value. Zero-padding, in my opinion, makes sense if you have input images with a pixel range [0, 255] or [0, 1] (after ...
6
votes
1answer
3k views

Understanding Leaky ReLU

I am reading Stanford's tutorial on the subject, and I have reached this part, "Training a Neural Network". So far so good. I understand pretty much everything. I would like to change the ReLU he is ...
6
votes
2answers
3k views

How does a Stacked AutoEncoder increases performance of a Convolutional Neural Network in image classification tasks

Stacked Auto-Endocer provides a version of raw data with more promising feature information, that can be used to train a classier with a specific context and find better accuracy than training ...
6
votes
2answers
8k views

Simple way to cluster histograms

I'm trying to cluster set of histograms. The histograms represent the frequencies of the distribution for a numbers from 1 to 5. The following figure shows two samples of my data. I have 10,000 ...
6
votes
1answer
2k views

How does FaceNet (Google's facerecognition) handles a new image?

I am currently researching in the facerecognition field. And I can not understand how the facenet algorithm handels a new image They use an euclidean space for image representation. Which means that ...
6
votes
3answers
4k views

use of t-test to compare performance of algorithms

I need a little bit guidance. I have to compare the classification performance of multiple algorithms using simple or paired t-test. Let's say I have four datasets (A,B,C) with training and test ...
6
votes
2answers
3k views

Understanding Median Frequency Balancing?

This question is with reference to semantic segmentation. According to the paper Predicting Depth, Surface Normals and Semantic Labels with a Common Multi-Scale Convolutional Architecture: we ...
6
votes
2answers
2k views

Why neural and convolutional neural network detect edges first?

There are many filters doing different tasks for us like edge filters, bar filters. Major filter banks are The Leung-Malik (LM) Filter Bank The Schmid (S) Filter Bank The Maximum ...
5
votes
3answers
10k views

How to calculate optimal zero padding for convolutional neural networks?

So formula for calculating the number of zero padding according to cs231n blog is : P = (F-1)/2 where P is number of ...
5
votes
2answers
205 views

Reference request - Computer Vision Book

What are the best books for obtaining a strong understanding of computer vision? From what I understand based on my undergraduate class, almost all current state-of-the-art computer vision is just ...
5
votes
2answers
5k views

Image reconstruction using compressed sensing

This question deals with an example of image reconstruction related to this other question on signal reconstruction. I have different issues in both examples but there could be an underlying factor. ...
5
votes
3answers
4k views

Convolution with a non-square kernel

So far I've only encountered convolution kernels which are square (ie, have the same rows as columns). Are there any cases in which a non-square kernel makes sense? If not, why?
5
votes
2answers
290 views

Ways of implementing Translation invariance

Is there any literature about the different ways translation invariance can be achieved when classifying images with Convolutional Neural Networks? Aside from using the structure of CNN, did anyone ...
5
votes
2answers
5k views

Difference between Mean/average accuracy and Overall accuracy

I just got confusion while reading the paper "Local Binary Pattern-Based Hyperspectral Image Classification With Superpixel Guidance". They mentioned that they repeated each experiment 10 times and ...
5
votes
1answer
5k views

Per Image Normalization vs overall dataset normalization

I am confused whether the standardization (subtract mean and divide by std) should be done per image basic or across the overall dataset. While overall dataset makes more sense, popular libraries like ...
5
votes
1answer
1k views

How is PCA used for 3D face reconstruction from 2D image?

I have learnt PCA in my data mining course and I know it is used for dimensionality reduction but I am confused about it how does it used for making 3D Morphable Model. Here is the paper link: Volker ...
5
votes
1answer
1k views

Computer vision algorithm that maps the positions of objects in 3D onto 2D image

Here is that I want to achieve: the input is a soccer image frame Then, I want my model to return a 2D model has following information: Where player's position in the field are given; and the player'...
5
votes
1answer
1k views

a simplified version of fully convolutional network

In the paper of fully convolutional networks semantic segmentation, the authors adopts up-sampling (de-convolutional network) to recover the feature maps with reduced dimensions (due to the multiple ...
5
votes
1answer
3k views

How do I classify images with non-rectangle shape with CNN?

How do I choose filter? Should I zero-padding the rest of the image and make it a rectangle? Or in a more specific case, while dealing with brain fMRI datas, the 3D models where brains are represented ...
5
votes
2answers
395 views

SVM with non-negative weights

An SVM classifier can be obtained by solving the following, $\arg\min \frac{1}{2}\|W\|_2^2 + C\sum_i \max(0, 1-y_i (W^T\mathbf{x}_i + b))$ where $W$ is the hyperplane (or weights), $b$ is the bias, $...
4
votes
1answer
8k views

Why is resnet faster than vgg

In Kaiming He's resnet presentation on slide 40 he says, "lower time complexity than VGG-16/19." Why is this the case, when resnet is much deeper?
4
votes
1answer
805 views

State of the art method for image captioning

Image captioning problem: Given an image, describe what is happening in the image. What is the state of the art work on image captioning?
4
votes
1answer
1k views

regarding the understanding of bottleneck unit of ResNet

In an article talking about ResNet, there has the following statement The second, the bottleneck unit, consists of three stacked operations. A series of 1x1, 3x3 and 1x1 convolutions substitute the ...
4
votes
1answer
626 views

IID in real life /Machine Learning - When is data truly IID? [duplicate]

In a course I am studying at Berkeley, some student said about a particular Dataset "Data is not iid" and the lecturer agreed with him. https://youtu.be/kl_G95uKTHw?list=...
4
votes
2answers
4k views

Pixel-wise classification on a large image using deep learning network

I am trying to classify every pixel on a large image (satellite image ~ 6000x4000 pixels) as belonging to one of the 4 classes:"Cloud", "Thin Cloud", "Clear", "Shadow." To that extent, I have taken ...
4
votes
2answers
1k views

Image processing with neural network [closed]

I am trying to learn how neural networks work on image recognition. I am confused on how to give input to neural network. Let's say I want to find (track) object in sequence of images, in particular ...

1
2 3 4 5
7