Questions tagged [computer-vision]

For questions related to computer vision, which is an interdisciplinary scientific field (which can e.g. use image processing techniques) that deals with how computers can be made to gain high-level understanding from digital images or videos. For example, image recognition (that is, the identification of the type of objects in an image) is a computer vision problem.

Filter by
Sorted by
Tagged with
4
votes
1answer
1k views

What do the words "coarse" and "fine" mean in the context of computer vision?

I was reading the well know paper Fully Convolutional Networks for Semantic Segmentation, and, throughout the whole paper, they talk use the term fine and coarse. I was wondering what they mean. The ...
1
vote
0answers
49 views

What is the reason for different learned features in upper and lower half in AlexNet?

I was reading AlexNet paper and the authors quoted the kernels on one GPU were "largely color agnostic," whereas the kernels on the other GPU were largely "color-specific." The upper GPU takes ...
1
vote
2answers
198 views

Why is the validation performance better than the training performance?

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 ...
1
vote
0answers
30 views

How to calibrate model's prediction given past images?

I want to predict how open is the mouth given a face image. It's a regression problem (0= mouth not open, 1=mouth completely open). And something between 0 and 1 is also allowed. ConvNet works fine ...
1
vote
1answer
50 views

Corner detection algorithm gives very high value for slanted edges?

I have tried implementing a basic version of shi-tomasi corner detection algorithm. The algorithm works fine for corners but I came across a strange issue that the algorithm also gives high values for ...
5
votes
2answers
337 views

Is the word "pose" used correctly in the paper "Matrix Capsules with EM Routing"?

In traditional computer vision and computer graphics, the pose matrix is a $4 \times 4$ matrix of the form $$ \begin{bmatrix} r_{11} & r_{12} & r_{12} & t_{1} \\ r_{21} & r_{...
2
votes
0answers
40 views

How exactly is equivariance achieved in capsule neural networks?

I have read quite a lot about capsule networks, but I cannot understand how the squashed vector would also rotate in response to rotation or translation of the image. A simple example would be helpful....
0
votes
1answer
41 views

Action recognition using video stream data

Recently, I am working on an action recognition project where my input data is from the video stream. I read some of the concepts like ConvLstm, Convolutional Lstm, etc. I am looking for someone who ...
1
vote
1answer
46 views

How well can a CNN distinguish an object from its class?

A convolutional neural network (CNN) can easily predict the class of an object in an image. Can a CNN distinguish the Pisa Tower from other buildings, or Hagia Sophia from other mosques easily? If ...
1
vote
0answers
142 views

Can we force the initial state of a neural network to produce an "unknown" class?

Has anyone investigated ways to initialize a network so that everything is considered "unknown" at the start? When you consider the ways humans learn, if something doesn't fit a class well enough, it ...
2
votes
2answers
110 views

What is the reasoning behind the number of filters in the convolution layer?

Let's assume an extreme case in which the kernel of the convolution layer takes only values 0 or 1. To capture all possible patterns in input of $C$ number of channels, we need $2^{C*K_H*K_W}$ filters,...
5
votes
3answers
1k views

Neural Network for Optical Mark Recognition?

I've created a neural net using the ConvNetSharp library which has 3 fully connected hidden layers. The first having 35 neurons and the other two having 25 neurons each, each layer with a ReLU layer ...
2
votes
0answers
28 views

Detect object in video and augment another video on top of it

I'm trying to detect an object in a video (with slight camera movement), and then augment another video on top of it. What is the simplest approach to do that? For instance, let's assume I have this ...
2
votes
0answers
29 views

How is visual attention mechanism different from a two branch convolutional neural network?

I am doing some research on the visual attention mechanism in remote sensing domain (where the features learnt from one layer are highlighted using the attention mask derived from another layer). From ...
4
votes
1answer
949 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
1answer
215 views

How does the math behind heat map filters work?

I am working on an app that generates heat/ thermal map given a picture. i have been able to get what i expected using python opencv builtin function ...
2
votes
0answers
53 views

Merge two different CNN models into one

I have 2 different models with each model doing a separate function and have been trained with different weights. Is there any way I can merge these two models to get a single model. If it can be ...
2
votes
1answer
1k views

How do you find the homography matrix given 4 points in both images?

I want to understand the process of finding a homography matrix given 4 points in both images. I am able to do that in python OpenCV, but I wonder how it works behind the scenes. Suppose I have ...
1
vote
0answers
36 views

An Encoder-Decoder based CNN to predict a tensor of points

So I have with me a data of rendered 2D images of a 3D object and along with that, I have the image projection coordinates (X, Y) of all the voxels that are in the ...
1
vote
1answer
34 views

Why do we set offset (0.5) in single shot detector?

In the paper SSD: Single Shot MultiBox Detector, under section 2.2 - (4), why do we add an offset of 0.5 to x, y in generating the anchor boxes across feature maps?
1
vote
0answers
25 views

How come a detection works after global average pooling 2D?

I use an off-the-shelf convolutional neural network, where at the end of the convolutional part, the depth of the last convolutional layer is expanded and then its 2D average is computed (such that ...
1
vote
1answer
54 views

Why are denser layers needed in computer vision neural nets?

Many neural net architectures for computer vision tasks use several convolutional layers and then several fully-connected (or dense) layers. While the reasons for using convolutional layers are clear ...
5
votes
1answer
754 views

Why is image recognition a key function of AI?

Image recognition, in the context of machine vision, is the ability of software to identify objects, places, people, writing and actions in images. Computers can use machine vision technologies in ...
2
votes
2answers
1k views

Could machine learning be used to measure the distance between two objects from a picture or live camera?

Could machine learning be used to measure the distance between two objects from a picture or live camera? An example of this is the measurement between the centre of each eye pupil. This area is ...
3
votes
0answers
168 views

Understanding the results of "Visualizing and Understanding Convolutional Networks"

I am trying to understand the results of the paper Visualizing and Understanding Convolutional Networks, in particular the following image: What are these 3x3 blocks and their 9 cells representing? ...
2
votes
0answers
39 views

Creating Dataset for Image Classification

I want to develop a CNN model to identify 24 hand signs in American Sign Language. I created a custom dataset that contains 3000 images for each hand sign i.e. 72000 images in the entire dataset. For ...
1
vote
1answer
247 views

What is "natural image domain"?

I see some papers use the term "natural image domain". I googled that but didn't find any explanation of it. I guess I understand the normal meaning of "natural image", such as the image people take ...
1
vote
0answers
49 views

What are some real-world products or applications that can be developed using GANs?

GANs have shown good progress across a wide variety of domains ranging from image translation, image generation, text to image synthesis, audio/video generation, image super-resolution and many more. ...
1
vote
1answer
328 views

What is a landmark in computer vision?

I guess I understand the concept of face detection, a technique specifies the location of multiple objects in the image, and draws bounding boxes on the target. The question is related to the concept ...
1
vote
1answer
41 views

Running 10 epochs on the Food-101 dataset

I’m currently working on the Food-101 dataset. I want to train a model that is greater than 85% accuracy for top-1 for the test set, using a ResNet50 or smaller network with a reasonable set of ...
3
votes
1answer
76 views

Before GAN, what are the commonly used techniques for image-to-image translation?

As per a post, image-to-image translation is a type of CV problem. I guess I understand the concept of image-to-image translation. I am aware that GANs(generative adversarial networks) are good at ...
1
vote
1answer
52 views

How to estimate the accuracy upper limit of any CNN model over a computer vision classification task

We are given a computer vision classification task, that is, a task that asks us to predict the category of an image over $n$ predefined classes (the so-called closed set classification problem). ...
2
votes
1answer
172 views

Is there any other rotated object detection datasets?

I have googled for a long time for rotated object detection datasets. Most of papers focused on rotated object detection using DOTA, HRSC2016 or coco text detection dataset. Some researcher also ...
2
votes
0answers
93 views

Why are conics important in computer vision?

The book Multiple View Geometry in Computer Vision by Richard Hartley and Andrew Zisserman talks about lines, points and conics. A conic is a curve described by a second-degree equation in the plane, ...
2
votes
0answers
56 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 ...
3
votes
1answer
175 views

How can I detect the frame from video streaming that contains a graffiti on city wall?

I am working on a graffiti detection project. I need to analyze data stream from a camera mounted sideways on a vehicle to identify graffiti on city walls and notify authorities with the single best ...
1
vote
0answers
41 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 ...
3
votes
1answer
51 views

What is the type of problem requiring to rate images on a scale?

I'm new to the topic, but I've used some off the shelf knowledge about computer vision for classifying images. For example, you can easily generate labels that can determine whether or not e.g. a ...
3
votes
1answer
84 views

Is there a way of computing a prominence score based on the prevalence of features in an image?

Is there any previous work on computing some sort of prominence score based on the prevalence of features in an image? For example, let's say I am classifying images based on whether or not they have ...
1
vote
0answers
84 views

What are the pros and cons of Keras, PyTorch and Caffe for computer vision?

I have tried to get the basic grasp of the following deep learning frameworks with python: Keras Pytorch Caffe However, I have lately noticed that people in the computer vision community care less ...
2
votes
0answers
29 views

Suitable algorithms for classifying terrain condition (asphalt, dirt etc) for motor vehicles

I am required to obtain data through a sensor located on the vehicle reading speed, vibration, roll and tilt, within a sample time, to classify the current road condition using machine learning for a ...
1
vote
0answers
28 views

Train an AI to infer accurate mathematical calculations by simply “looking” at images of shapes/objects

I’d like to build a model that has an understanding of geometry, where it can be applied to question and answering system. Specifically, it would be nice if it could determine the volume of an object ...
0
votes
0answers
81 views

Intuition behind single-shot object detection

Is there a good way to understand how single-shot object detection works? The most basic way to do detection is use a sliding-window detector and look at the output of the NN to detect if a class is ...
0
votes
1answer
153 views

YOLOv3 Model Structure: Why is filters = (classes + coords + 1) * num?

Here's a tutorial about doing custom training of YOLO (Darknet): https://medium.com/@manivannan_data/how-to-train-yolov3-to-detect-custom-objects-ccbcafeb13d2 The tutorial guides how to set values in ...
1
vote
2answers
789 views

What are the reasons behind slow YOLO training?

I'm testing out YOLOv3 using the 'darknet' binary, and custom config. It trains rather slow. My testing out is only with 1 image, 1 class, and using YOLOv3-tiny instead of YOLOv3 full, but the ...
1
vote
0answers
80 views

Weird border artifacts when training a CNN

I've been trying to use this DeepLabv3+ implementation with my dataset (~1000 annotated images of the same box, out of the same video sequence): https://github.com/srihari-humbarwadi/...
1
vote
1answer
194 views

What is "temporal depth"?

I need some explanation about the following paragraph (page 3) from the paper A Novel Approach for Robust Multi Human Action Detection and Recognition based on 3-Dimentional Convolutional Neural ...
1
vote
1answer
35 views

Interpolating image to increase resolution before feeding it to a neural network

Interpolation is a common way to make an image fit the right input shape for a neural network. But is there any point in using interpolation to make it easier for the network to learn? I assume ...
2
votes
0answers
25 views

What are the current tools and techniques for image segmentation in order of pragmatism?

To explain what I mean I'll depict the two extremes and something in the middle. 1) Most pragmatic: If you need to just segment a few images for a design project, forget AI. Go into Adobe Photoshop ...
1
vote
0answers
17 views

How can a new metric applied for humans causing danger on railtracks?

I am writing myself and was thinking about, what kind of metric can be applied to measure the "dangerousness" of a human being on a railtrack? For example detecting if a human is running on the rails? ...

1
5
6
7 8 9