Questions tagged [computer-vision]
Questions about algorithms for processing and analyzing images and videos.
235
questions
-1
votes
0answers
16 views
Machine learning methods
In terms of a deep neural network, which activation will perform better between ReLU and sigmoid? Does it depend on the data or is one always better than the other? Are there any associated costs?
0
votes
0answers
8 views
Bi-quadratic Interpolation
If bilinear interpolation is faster (uses 4 data points) but less accurate than bicubic interpolation, which uses 16 data points, why is there no "bi-quadratic" interpolation, which would use 9 data ...
0
votes
0answers
11 views
Measuring distance in arbitrary scale point cloud
I have one point cloud(point cloud of human feet) of arbitrary scale. If I have another point cloud(of human feet) with dimensions known, can I get the dimensions of a first point cloud? How will ...
0
votes
0answers
8 views
Using stereo vision for obstacle avoidance on quadcopter?
I am basically an electrical engineer. I plan to use stereo vision for obstacle avoidance (and possible further development of object recognition). Any efficient libraries for depth mapping from ...
3
votes
0answers
25 views
Partial derivatives of bilinear Sampling
I'm going over the paper Spatial Transformer Networks (link). On section 3.3 they introduce bilinear sampling kernel (Eq. 5)
$$
V_{i}^{c}=\sum_{n}^{H} \sum_{m}^{W} U_{n m}^{c} \max \left(0,1-\left|x_{...
2
votes
2answers
45 views
How to detect number of specific object in an image and it's diameter(size)?
I have two images of IRON RODS (i.e only two images) and I want to detect how many rods are there in each image and get the diameter of each rod in a image?
I don'...
1
vote
0answers
18 views
RGBD alignment without explicit transformation between RGB and depth
I have a set of RGB-D scans of the same scene, corresponding camera poses, intrinsics and extrinsics for both color and depth cameras. RGB's resolution is 1290x960, depth's is 640x480.
Depth is ...
4
votes
1answer
111 views
Difference between Image Analysis and Computer Vision
I have searched the web and I have only found the differences between Image Processing and Computer Vison (also in this website too). In addition, sometimes Image Analysis is confused with Image ...
0
votes
0answers
26 views
Graph neural network
I'm trying to build GNN model that classify images , the first step is to model each image with graph , each node represents one pixel , now how can I define the edges in my case ? does the spatial ...
1
vote
0answers
55 views
Construct 3D Point Cloud from 2D Depth Map
I have the following image
from which I create a depth map which looks as follows
From this 360 degree Depth Map I would like to create a Cartesian point cloud. Are there any set methods or ...
3
votes
0answers
16 views
Using colors in feature detection
Is there any feature detector or extractor that uses color information?
Detectors like SURF and SIFT and many others work with grayscale. How do you think, may be transform can be sued to generate a ...
2
votes
0answers
37 views
Do image artefacts reveal the original image orientation?
An image is taken and then rotated by an arbitrary angle and cropped. Let us assume that the original image is rather meaningless and is not aligned in any way. Given the rotated image, is it ...
0
votes
1answer
48 views
Computer Vision Techniques to find Slope of beach in the 2D Images?
What I am planning to do is to calculate the height and estimate the slope of the segmented Object. The camera will be static and the Object of Interest is the slope of the beach.
I am finding harder ...
0
votes
0answers
19 views
How to transform an arbitrary graph into a fixed vector representation?
Actuality I work in computer vision, specifically on a problem known as "scene graph modeling." This problem aims to convert an image $I$ in a graph $G=(V,E)$ where the nodes $V$ represent the objects ...
1
vote
0answers
9 views
Descriptors in key-features during mapping
This question is related to this question.
In SLAM-type applications the feature detectors like SIFT are used for finding keypoints in the images to be matched. The candidate keypoints are matched by ...
2
votes
1answer
34 views
How are image processing and computer vision related to artificial intelligence.?
We have often heard both terms image processing and computer vision in AI sessions
But i am confused, how they both are related to artificial intelligence?
1
vote
0answers
17 views
Where is the epipole if one camera center is not in view of the other?
In the book multiple view geometry, the epipole is defined as follows:
The epipole is the point of intersection of the line joining the
camera centres (the baseline) with the image plane. ...
0
votes
0answers
10 views
Using Particle Filters to find optimal 3D rotation
There exists a set of points in a 3D euklidean space. A camera views that point set and that camera is aligned with the world coordinate system.
The center of gravity of the point cloud is in the ...
1
vote
0answers
13 views
Does the illumination color (red/blue/green) has any effect over the disparity map generated from a stereo image?
I have a working model that creates a disparity map using stereoscopic image processing using opencv. I am thinking of improving the disparity map by using illumination and stuck with different ...
1
vote
1answer
25 views
How to calculate point in 3D space from multiple perspective
I have multiple camera in different points.
I have their position and rotation as $(x,y,z)$ , $(\alpha,\beta,\gamma)$ or $( roll, pitch, yaw)$ .
And I have output like this :
Feed from camera-1, I ...
0
votes
0answers
15 views
maximum tensor rank for computer vision tasks
Most computer vision tasks do not require 4+ ranked tensors.
Are there any use cases for tensors with ranks more than 4 in computer vision? Are there any use cases for deep learning in general?
1
vote
0answers
40 views
What is the common way to approximate contour with quad?
I'm trying to find quads on stream from my webcam. cv2.HoughLinesP sometimes returns more lines then needed. How can I approximate them to contour of exactly 4 points that fits returned result best? I'...
3
votes
2answers
61 views
How to detect facial landmarks - using haar and other way?
I have done some projects using haar cascade(Ex.- Face-detection,Eye inside face detection,shoe detection etc), I also have some knowledge about it. I have trained my own models learning from youtube, ...
3
votes
0answers
55 views
Calculating depth mask from different lighting
I have a object which is static, the camera is static and light source is moving. How can the depth mask be calculated ?
Concept is to use - calculate height from shadow length
Lets imagine a have ...
3
votes
0answers
15 views
impact of the single pixel in disparity
Disparity essentially means the distance of a corresponding pixel in another image ( in pixels). This is clear and understood. When I looked after how the depth is related to disparity, then I found ...
1
vote
1answer
52 views
Triangulation algorithm for an arbitary amount of points?
I have multiple cameras that records the 2D pixel coordinates of the features each camera sees. After some searching, I came across this paper with a method to calculate the 3D coordinates of the ...
2
votes
2answers
131 views
Computer Vision algorithm to tell if camera is moving?
I'm looking for a computer vision algorithm or method that can tell if the camera is moving in a video. Or maybe an alternate way of telling if the background is moving. I have a lot of videos and ...
0
votes
0answers
15 views
Confused between different Structure from Motion pipelines
What is the difference between Global SfM, Incremental SfM, Sequential SfM, Progressive SfM and VSLAM (Visual Simultaneous Localization and Mapping). Is sequential SfM the same as VSLAM ? I have tried ...
1
vote
1answer
68 views
Image registration using gradient descent
I have a target image $f(x,y)$ (where $x \in [0, 250]$ and $y \in [0,300]$), and a source image $g(x,y)$
I want to align $g$ to $f$ using the transformation :
$$\Psi(x,y;t_x, t_y, \theta) = \begin{...
1
vote
0answers
48 views
Why is sparsity useful in dictionary learning?
What is the particular reason for which we use sparse matrix X in dictionary learning equation Y=DX where Y is the signal matrix, D is dictionary and X is the coefficient matrix?
2
votes
2answers
39 views
identify abrupt frame changes in a video
The frames of live streams or videos, in general, do not flow smoothly. A sports game would have multiple cameras recording the game and hence, there will be camera switches in the stream. I am ...
1
vote
0answers
29 views
Finding the linear mapping between homogeneous coordinates of affine camera
If I have an affine camera with a projection relationship governed by:
\begin{equation}
\begin{bmatrix}
x & y
\end{bmatrix}^T = A
\begin{bmatrix}
X & Y & Z
\end{bmatrix}^T + b
\end{...
0
votes
1answer
80 views
Camera calibration without grid
Ususally a grid pattern (circles or squares) is used for camera calibration. Features (circles or corners) are detected in a function (e.g. findCirclesGrid in OpenCV) and passed to the a camera ...
-2
votes
1answer
63 views
what the mean and median filters with an n x n kernel (neighborhood) do to the grey-levels of the image? [closed]
sorry i haven't found a good title for these three questions
i'm wondering if someone could give me explaination with proof (if possible ) of these three questions :
Given a grey-level image stored ...
0
votes
1answer
31 views
How do we fix area of detectors in object detection?
I have gone through various articles on medium and also some from other sites trying to understand SSD. I am able to figure out most of the things from articles except this one.
They always say that ...
1
vote
0answers
20 views
3D registration for reconstructed surface
I use Meshroom (based on the framework AliceVision) to reconstruct a scene based on photographs (e.g. a bottle on a table in a room). The software captures the bottle, but also the table and the whole ...
1
vote
0answers
45 views
Algorithm to determine paths that define shape of connected rectangles
Say you have a bunch of rectangles like these:
They get organized into 3 distinct final shapes like this:
As you can see, there are a few "complete contours", as in here:
The way you figure out the ...
0
votes
0answers
76 views
What is the practical limit to how many object classes you can detect with Faster RCNN?
I am trying to follow this tutorial where the Faster-RCNN-Inception-V2-COCO model from TensorFlow's model zoo is used to detect playing cards. I was wondering what is the practical limit to the number ...
1
vote
1answer
102 views
When estimating camera motion from the fundamental matrix, why can translation only be recovered up to scale using epipolar geometry?
I was wondering once you have recovered the fundamental matrix F such that xFx' = 0. If you try to recover translation and rotation from F, I am told that this is possible only up to scale? Why so?
1
vote
0answers
37 views
Stereo-Camera: How do Bundle Adjustment and projective ambiguity fit together?
In the book Multiple View Geometry in Computer Vision Hartley and Zisserman write that given a stereo image pair, the viewed scene can only be reconstructed under projective ambiguity, i.e. is a ...
1
vote
0answers
67 views
Transform Image Point to Azimuth and Elevation
I'm working on an object tracker where I need to report the azimuth and elevation of targets. Both values should be scaled from degrees to unsigned 32-bit integers. For now I can ignore lens ...
1
vote
0answers
49 views
How to find corresponding points between two photos when there's a repeated pattern?
Suppose that I have two photos, I1 and I2 and I want to find a set of corresponding points between the two photos to define a homography between them. There are many algorithms for this purpose, ...
1
vote
0answers
28 views
Removing fish-eye effect
The DNG specification (page 87) describes a simple algorithm for converting a "fish-eye" photo into a regular photo (just several lines of basic geometry).
The transformation is configured using six ...
2
votes
1answer
92 views
Do white box methods for feature extraction exist for facial recognition?
Part of something I'm working on (or rather begin working on) involves white box approaches to facial recognition. I previously worked on something for character recognition using scene segmentation ...
1
vote
0answers
15 views
Plenoptic function meaning of dx
I'm trying to understand this paper on computer vision (http://persci.mit.edu/pub_pdfs/elements91.pdf) and it talks about building a periodic table of things we can get from the plenoptic function. It ...
1
vote
1answer
63 views
Image selection for Object detection
I'm relatively new and am still learning the basics. I've used NVIDIA DIGITS in the past, and am now looking at Tensorflow. While I've been able to fumble my way around creating some models for a few ...
1
vote
1answer
45 views
After stereo image rectification, why does the relation between disparity and depth still apply?
The relation between disparity and depth is derived based on the assumption that the two cameras are parallel.
Image rectification does not make this assumption valid.
But I see in many university ...
1
vote
0answers
43 views
Is there ongoing research on Markov Chain Monte Carlo algorithms (other than Metropolis-Hastings) for use in ray tracing? [closed]
Metropolis Light Transport is an application of the Metropolis-Hastings algorithm. Its variants differ only in the used "mutation strategy".
Is there ongoing research on other Markov Chain Monte ...
2
votes
1answer
46 views
Hough Transform, Computer Vision
There is some thing i dont understand about Hough Transform in polar coordinates.
From wikipedia article:
"The simplest case of Hough transform is detecting straight lines. In general, the straight ...
1
vote
1answer
52 views
Comparing variable image to static image
What are Levenshtein distance algorithms, but for images?
I will have one static image (the target: what it should look like) and one variable image (actual: what the user has created)
For example,
...
