I have a 512x512 grayscale image (or MultiArray) which is the output of a CoreML depth estimation model.
In Python, one can use Matplotlib or other packages to visualise grayscale images in different colormaps, like so:
Grayscale
Magma
[Images from https://ai.googleblog.com/2019/08/turbo-improved-rainbow-colormap-for.html]
I was wondering if there was any way to take said output and present it as a cmap in Swift/iOS?
If you make the model output an image, you get a CVPixelBuffer object. This is easy enough to draw on the screen by converting it to a CIImage and then a CGImage.
If you want to draw it with a colormap, you'll have to replace each of the grayscale values with a color manually. One way to do this is to output an MLMultiArray and loop through each of the output values, and use a lookup table for the colors. A quicker way is to do this in a Metal compute shader.
Related
I trained an image to image translation model on pytorch and the input and output images are in CIELAB color space. How do I convert this to an RGB image? Simply converting the image causes some sort of clipping and produces white patches.
out=model.forward(x)
out=torch.squeeze(out)
out=out.permute(1,2,0)
out=torch.from_numpy(out.data.numpy())
plt.imshow(out)
This doesn't produce white patches however I cant use OpenCV and convert it to RGB as the values are in range 0-1.
Now if I convert the tensor to a PIL image and then convert to RGB(0-255) some sort of clipping occurs and produces white patches which are even visible before converting to RGB
out=model.forward(x)
out=torch.squeeze(out)
out=np.asarray(transforms.ToPILImage()(out))
plt.imshow(out)
The white patches after using out=cv2.cvtColor(out, cv2.COLOR_LAB2RGB) to convert
How can I properly convert the CIELAB image to RGB?
I am trying to understand the jpeg compression algorithm. If I have a 3-channel color image, do I have to take 3 different Discrete cosine transform (DCT) and quantize for each channel? And after taking inverse DCT, will the result be an jpeg image?
If I have a 3-channel color image, do I have to take 3 different Discrete cosine transform (DCT) and quantize for each channel?
Yes, except that the color values are normally converted from RGB to YCbCr first.
Then you have to do run-length compression and Huffman coding on the resulting values. The DCT alone negatively compresses.
I have a problem with normalization.
Let me what the problem is and how I attempt to solve it.
I take a three-channel color image, convert it to grayscale and apply uniform or non-uniform quantization and the same thing.
To this image, I should apply the normalization, but I have a problem even if the image and grayscale and always has three channels.
How can I apply normalization having a three-channel image?
Should the min and the max all be in the three channels?
Could someone give me a hand?
The language I am using is processing 2.
P.S.
Can you do the same thing with a color image instead use a grayscale image?
You can convert between the 1-channel and 3-channel representations easily. I'd recommend scikit-image (http://scikit-image.org/).
from skimage.io import imread
from skimage.color import rgb2gray, gray2rgb
rgb_img = imread('path/to/my/image')
gray_img = rgb2gray(rgb_image)
# Now normalize gray image
gray_norm = gray_img / max(gray_img)
# Now convert back
rgb_norm = gray2rgb(gray_norm)
I worked with a similar problem sometime back. One of the good solutions to this was to:
Convert the image from RGB to HSI
Leaving the Hue and Saturation channels unchanged, simply normalize across the Intensity channel
Convert back to RGB
This logic can be applied accross several other image processing tasks, like for example, applying histogram equalization to RGB images.
I need to quantize the RGB values to 29 uniform color dictionary. I used rgb2ind(image,29) in Matlab.
So, is there any function or efficient way to quantize the image color in OpenCV?
(I need to quantize the image color because i want to get a 29-sized histogram of color)
You will have to make your own. I can reccomend using HSV instead of RGB (you can convert RGB to HSV with opencv). Once the image is converted, you can then simply use 29 ranges for the H value.
EDIT: I saw this answer might be a bit vague for those who have little experience in computer vision. This question gives a lot more information about the difference between HSV and RGB and why this is usefull.
I am using SURF features in OpenCV where the input images are converted to GRAY image.
cvtColor(object, object, CV_RGB2GRAY);
When I went through the documentation of OpenSURF I realised that its not in grayscale.
My confusion is that can we apply SURF to any image formats (YUV, HSV, RGB) or we have to change and modify the program to achieve that?
Most feature detectors work on greyscale because they analyse the patterns of edges in the image patch. You can run SURF on any single colour channel from the colour formats you mention i.e. You can run it on Y, U or V from YUV images, or on H, S or V from HSV images. Not sure how OpenSURF treats this, but they must be using the greyscale image internally.
Like OpenCV if you given an image to OpenSURF that is not single channel, it calls cvtColor(src, dst, CV_BGR2GRAY). If you pass either a 3 channel image in a YUV, HSV, Lab etc, things will go horribly wrong because the image will have an inappropriate color conversion applied..