Depth-RGB alignment

[linwanml]

Is there a method to read the depth value from the depth video by using python and opencv?

[r9y9]

Assuming you are talking about reading depth values from a pre-recorded depth sequence, I think you can just use opencv’s VideoCapture.

[linwanml]

Yes, I have already read the value with opencv. The depth intensity in opencv is 0-255. Is there a transform equation to map each depth intensity(0-255) into millimeter values?

[r9y9]

That depends on how you saved depth data. For example, assuming the maximum depth in milli meter was 4500, If you scaled the depth values by 255/4500 and saved it as 8bit integers, then you can recover depth by multiplying 4500/255.

[linwanml]

Ok, thank you for the answers. I have one more question, and then I will close the issue.
Are there some documents, links or papers that I can have a look about these information?