[L515] align offset

[wangsff]

---

Required Info
Camera Model	{L515
Firmware Version	01.05.01.03
Operating System & Version	win10
Platform	PC
SDK Version	realsense viewer v2.38.1
Language	python }

Issue Description

As I mention on #7556 , use your align code has come up the "strange duplication region", so I want to align the color map to depth map with another menthod.
Here is code I use.

import pyrealsense2 as rs
import numpy as np
import cv2

config = rs.config()
config.enable_stream(rs.stream.depth, 1024, 768, rs.format.z16, 30)
config.enable_stream(rs.stream.color, 1280, 720, rs.format.bgr8, 30)

pipeline = rs.pipeline()

pipe_profile = pipeline.start(config)

frames = pipeline.wait_for_frames()
depth_frame = frames.get_depth_frame()
color_frame = frames.get_color_frame()
# Intrinsics & Extrinsics
depth_intrin = depth_frame.profile.as_video_stream_profile().intrinsics
color_intrin = color_frame.profile.as_video_stream_profile().intrinsics
depth_to_color_extrin = depth_frame.profile.get_extrinsics_to(color_frame.profile)
color_to_depth_extrin = color_frame.profile.get_extrinsics_to(depth_frame.profile)
colorizer = rs.colorizer(2)
colorizer_depth = np.asanyarray(colorizer.colorize(depth_frame).get_data())
cv2.imwrite(r'C:\Users\DELL\Desktop\depth.png', colorizer_depth)
# color_image = np.asanyarray(color_frame.get_data())
# cv2.imwrite(r'C:\Users\DELL\Desktop\color.png', color_image)

print ("\n Depth intrinsics: " + str(depth_intrin))
print ("\n Color intrinsics: " + str(color_intrin))
print ("\n Depth to color extrinsics: " + str(depth_to_color_extrin))

# Depth scale - units of the values inside a depth frame, i.e how to convert the value to units of 1 meter
depth_sensor = pipe_profile.get_device().first_depth_sensor()
# depth_scale = depth_sensor.get_depth_scale()
depth_scale = depth_sensor.get_option(rs.option.depth_units)
depth_image = np.asanyarray(depth_frame.get_data())
print ("\n\t depth_scale: " + str(depth_scale))
color_image = np.asanyarray(color_frame.get_data())

height = depth_image.shape[0]
width = depth_image.shape[1]
aligned_color = np.zeros((height, width, 3))
height_color = color_image.shape[0]
width_color = color_image.shape[1]

for v in range(width_color):
    for u in range(height_color):

        color_pixel = [v, u]
        color_point = rs.rs2_deproject_pixel_to_point(color_intrin, color_pixel, 1)
        depth_point = rs.rs2_transform_point_to_point(color_to_depth_extrin, color_point)
        depth_pixel = rs.rs2_project_point_to_pixel(depth_intrin, depth_point)
        # depth_pixel = rs.rs2_project_color_pixel_to_depth_pixel(depth_frame.get_data(), depth_scale,
        #     0.11, 1.0, depth_intrin, color_intrin, depth_to_color_extrin, color_to_depth_extrin, color_pixel)
        u_depth = int(round(depth_pixel[1]))
        v_depth = int(round(depth_pixel[0]))
        if u_depth < 0 or u_depth > height-1 or v_depth < 0 or v_depth > width-1:
            pass
        else:
            aligned_color[u_depth][v_depth][0] = color_image[u][v][0]
            aligned_color[u_depth][v_depth][1] = color_image[u][v][1]
            aligned_color[u_depth][v_depth][2] = color_image[u][v][2]
cv2.imwrite(r'C:\Users\DELL\Desktop\aligned_color.png', aligned_color)

And the aligned color map seems not to align very well.

aligned color map

origin depth map

You can see the aligned color map downward shift to the correct position.
How can I solve it? Thanks in advance!

[cdb0y511]

Hi, @wangsff
This may be related to #7335
The Inverse-Brown-Conrady model may not be accurate.

[RealSenseSupport]

FYI...

#7335 (comment)

Thanks

[RealSenseSupport]

Hi,

Will you be needing further help with this?

If we don’t hear from you in 7 days, this issue will be closed.

Thanks