15
votes
How to generate stereo image pair from a stationary mono camera?
You can't. You have video of the scene from a single vantage point. Without depth information, you can't infer what the scene would look like from another vantage point.
D.W.♦
- 156k
14
votes
What is the difference between luma and luminance?
There are actually three related terms:
Luminance is a physical measure which represents the luminous intensity per unit area of light travelling in some direction. The units are candela per square ...
- 21.7k
13
votes
2D convolution: Flipping the kernel?
If you do not flip the kernel, you simply obtain a different operation that is called cross correlation. When the filter is symmetric, like a Gaussian, or a Laplacian, convolution and correlation ...
- 496
9
votes
How to generate stereo image pair from a stationary mono camera?
The terms you probably want to google for are "inferring depth maps". Just like your brain tricks you into seeing 3d if you close one of your eyes, you can heuristically recover depth maps from single ...
- 5,931
7
votes
Google DeepDream Elaborated
The idea of DeepDream is this: pick some layer from the network (usually a convolutional layer), pass the starting image through the network to extract features at the chosen layer, set the gradient ...
- 640
7
votes
How does a predictive coding aid in lossless compression?
Huffman coding, as usually applied, only considers the distribution of singletons. If $X$ is the distribution of a random singleton, then Huffman coding uses between $H(X)$ and $H(X)+1$ bits per ...
- 275k
6
votes
What is the difference between the fundamental matrix and the essential matrix?
I want to add one thing to @Dima's answer, just so things are absolutely clear:
$E = (K')^TFK$ (note the transpose)
You can see how this also captures that the fundamental matrix takes image ...
- 61
5
votes
Accepted
High Dimensional Spaces for Images
From one perspective, a picture is a 2D image, because it has height and width.
But from a machine learning perspective, we can think of a picture as a point in a high-dimensional space. In ...
D.W.♦
- 156k
5
votes
Accepted
Hausdorff distance between two binary images according to distance maps
So I finally managed to do it, here's the answer for future reference.
The idea is not to compute the distance map of the image, but of its complement.
Here's a example to better understand :
You ...
- 201
4
votes
Accepted
Meaning of topological distance between 2 pixels
Nonlinear Signal and Image Processing: Theory, Methods, and Applications defines topological distance as follows. First, you have to define when two pixels are neighbors. The book offers two ...
- 275k
4
votes
Slicing clusters of non-black pixels from an image into separate images
It seems that you're just trying to find the connected components of a graph. In this case, the vertices are the non-black pixels and there's an edge between two pixels if they're at distance at most&...
- 81.4k
4
votes
Accepted
Computer vision: object detection with labels that are single coordinates
The state of the art in such problems is done these days via deep neural networks. Among others, two popular and recent approaches for solving the problem of detection and localization of objects are ...
- 398
4
votes
Accepted
Rearrange pixels to maximize visually distinguishable pattern
This will not be an algorithm in the usual sense. It's a bunch of heuristics that you may find useful. The main difficulty comes from the undefined nature of the phrase "visually similar".
If I ...
- 1,787
4
votes
Accepted
Summed-area table for circular range queries
The best way is to approximate circle with hexagon or octagon as desciribed in expired patent US 6507676 B1. Further improvement is to add rectangles to octagon to get more circular shape or like ...
- 9,425
3
votes
What is a visual bag of words and how is it implemented?
This is the original paper:
Gabriella Csurka, Christopher R. Dance, Lixin Fan, Jutta Willamowski, and Cédric Bray. Visual Categorization with Bag of Keypoints. Workshop on Statistical Learning in ...
- 578
3
votes
Extracting features for texture classification
If it is important for you to get a high accuracy, then use a convolutional neural network (ConvNet). These ConvNets hold the state of the art for most visual recognition tasks. If your training set ...
- 131
3
votes
Accepted
What is the appropriate means of bias field in image processing?
Bias field is magnetic based inhomogenity in results of MRI, which comes from magnetic settings, patient position etc.
This term has no standard definition in terms of image processing.
Essentialy ...
- 9,425
3
votes
Accepted
Which algorithm for counting the occurrences of a certain pattern (spots) in an image?
The best way to get an orientation is to read a textbook on computer vision or image processing. There are lots of techniques known, and that's often the best way to get an introduction to a broad ...
D.W.♦
- 156k
3
votes
Reconstructing a screen of permuted pixels
One approach is to exploit fact that if two pixels are adjacent, then their intensities will be highly correlated: the intensity of the first pixel (as a time series) will be highly correlated to the ...
D.W.♦
- 156k
3
votes
Accepted
How are the number of bytes less than the number of pixels in an image?
A colour image is typically digitalized using 256 levels for each of the 3 RGB channels. That gives 3 bytes per pixel. The trick to attain smaller file size is to apply some compression, to take ...
- 256
3
votes
Slicing clusters of non-black pixels from an image into separate images
You are probably looking for something like image segmentation.
- 5,931
3
votes
Accepted
Applicability of Machine Learning on Images.
It looks plausible that this is the kind of problem machine learning will be good at. I suspect you might be able to achieve much higher accuracy, but the only way to know for sure is to try it.
For ...
D.W.♦
- 156k
3
votes
Algorithms to correct misspelled word?
Standard OCR software already incorporates this idea: it has a corpus of words, along with their frequency of occurrence, and uses this to correct errors.
How does it work? OCR doesn't just output ...
D.W.♦
- 156k
3
votes
Convolutional Neural Network with constant kernels
A fully connected layer is able to classify images better than random. If you make the kernels constant and there is still information in the result, the network is essentially only a FCN operating on ...
- 2,350
3
votes
Accepted
Given a low resolution video, is it possible to create a higher resolution image
It heavily depends on the setting, accuracy and goal, but shortly, yes, even considering practical application.
Theoretically of course it is possible, with various (or used together) techniques like ...
- 9,425
3
votes
Sample K representative frames within a video
The question misses a lot of detail, so I will try to make an educated guess.
I don't know of any specific algorithm for the task you are trying to achieve, but the first step towards your solution ...
- 66
3
votes
Accepted
How to prove that the erosion of a convex set by a convex set will result in a convex set?
(Convexity is closed under intersection) Given a family of convex set $C_i$, where $i\in I$ for some index set $I$, then $\cap_{i\in I}C_i$ is convex.
In other words, the intersection of convex sets ...
- 38.6k
3
votes
Accepted
what is meant by support of a filter?
Many filters, particularly linear time-invariant ones, are represented by a kernel function. The output of a filter at a point is calculated by a weighted average of all the points of the input where ...
Only top scored, non community-wiki answers of a minimum length are eligible