How do you convert indexed color image numpy array to RGB color-space?

Question

I would like to convert indexed color image array to RGB color-space. I know I can achieve that if I have PIL read image like following.

with Image.open(png_image_path) as img:
    rgb_image = np.asarray(img.convert('RGB'))

I wanna do the same thing from numpy array rather than the image object read by PIL like following.

image = Image.fromarray(indexed_image).convert('RGB')

But it doesn't simply work the way I want. Any suggestion will be helpful to me! Thank you for taking your time.

Answer 1

You can do this:

scr = numpy.frombuffer(screen_buffer, dtype=numpy.uint8)
pal = numpy.frombuffer(palette_buffer, dtype=numpy.uint8).reshape((256, 3))

# Convert indexed color to RGB
arr = pal[scr]

data = arr.astype(numpy.uint8).tobytes()