Monodepth Estimation with OpenVINO

This tutorial is also available as a Jupyter notebook that can be cloned directly from GitHub. See the installation guide for instructions to run this tutorial locally on Windows, Linux or macOS. To run without installing anything, click the launch binder button.

Binder Github

This tutorial demonstrates Monocular Depth Estimation with MidasNet in OpenVINO. Model information can be found here.



What is Monodepth?

Monocular Depth Estimation is the task of estimating scene depth using a single image. It has many potential applications in robotics, 3D reconstruction, medical imaging and autonomous systems. This tutorial uses a neural network model called MiDaS, which was developed by the Embodied AI Foundation. See the research paper below to learn more.

R. Ranftl, K. Lasinger, D. Hafner, K. Schindler and V. Koltun, “Towards Robust Monocular Depth Estimation: Mixing Datasets for Zero-shot Cross-dataset Transfer,” in IEEE Transactions on Pattern Analysis and Machine Intelligence, doi: 10.1109/TPAMI.2020.3019967.



import sys
import time
from pathlib import Path

import cv2
import matplotlib.pyplot as plt
import numpy as np
from IPython.display import (
from openvino.runtime import Core

from notebook_utils import load_image


MODEL_FILE = "model/MiDaS_small.xml"

model_xml_path = Path(MODEL_FILE)


def normalize_minmax(data):
    """Normalizes the values in `data` between 0 and 1"""
    return (data - data.min()) / (data.max() - data.min())

def convert_result_to_image(result, colormap="viridis"):
    Convert network result of floating point numbers to an RGB image with
    integer values from 0-255 by applying a colormap.

    `result` is expected to be a single network result in 1,H,W shape
    `colormap` is a matplotlib colormap.
    cmap =
    result = result.squeeze(0)
    result = normalize_minmax(result)
    result = cmap(result)[:, :, :3] * 255
    result = result.astype(np.uint8)
    return result

def to_rgb(image_data) -> np.ndarray:
    Convert image_data from BGR to RGB
    return cv2.cvtColor(image_data, cv2.COLOR_BGR2RGB)

Load the Model

Load the model in OpenVINO Runtime with ie.read_model and compile it for the specified device with ie.compile_model. Get input and output keys and the expected input shape for the model.

ie = Core()
ie.set_property({'CACHE_DIR': '../cache'})
model = ie.read_model(model_xml_path)
compiled_model = ie.compile_model(model=model, device_name=DEVICE)

input_key = compiled_model.input(0)
output_key = compiled_model.output(0)

network_input_shape = list(input_key.shape)
network_image_height, network_image_width = network_input_shape[2:]

Monodepth on Image

Load, resize and reshape input image

The input image is read with OpenCV, resized to network input size, and reshaped to (N,C,H,W) (N=number of images, C=number of channels, H=height, W=width).

IMAGE_FILE = "../data/image/coco_bike.jpg"
image = load_image(path=IMAGE_FILE)

# Resize to input shape for network.
resized_image = cv2.resize(src=image, dsize=(network_image_height, network_image_width))

# Reshape the image to network input shape NCHW.
input_image = np.expand_dims(np.transpose(resized_image, (2, 0, 1)), 0)

Do inference on the image

Do inference, convert the result to an image, and resize it to the original image shape.

result = compiled_model([input_image])[output_key]

# Convert the network result of disparity map to an image that shows
# distance as colors.
result_image = convert_result_to_image(result=result)

# Resize back to original image shape. The `cv2.resize` function expects shape
# in (width, height), [::-1] reverses the (height, width) shape to match this.
result_image = cv2.resize(result_image, image.shape[:2][::-1])

Display monodepth image

fig, ax = plt.subplots(1, 2, figsize=(20, 15))

Monodepth on Video

By default, only the first 100 frames are processed in order to quickly check that everything works. Change NUM_FRAMES in the cell below to modify this. Set NUM_FRAMES to 0 to process the whole video.

Video Settings

# Video source: (Public Domain)
VIDEO_FILE = "../data/video/Coco Walking in Berkeley.mp4"
# Number of seconds of input video to process. Set `NUM_SECONDS` to 0 to process
# the full video.
# Set `ADVANCE_FRAMES` to 1 to process every frame from the input video
# Set `ADVANCE_FRAMES` to 2 to process every second frame. This reduces
# the time it takes to process the video.
# Set `SCALE_OUTPUT` to reduce the size of the result video
# If `SCALE_OUTPUT` is 0.5, the width and height of the result video
# will be half the width and height of the input video.
# The format to use for video encoding. The 'vp09` is slow,
# but it works on most systems.
# Try the `THEO` encoding if you have FFMPEG installed.
# FOURCC = cv2.VideoWriter_fourcc(*"THEO")
FOURCC = cv2.VideoWriter_fourcc(*"vp09")

# Create Path objects for the input video and the result video.
output_directory = Path("output")
result_video_path = output_directory / f"{Path(VIDEO_FILE).stem}_monodepth.mp4"

Load the Video

Load the video from a VIDEO_FILE, set in the Video Settings cell above. Open the video to read the frame width and height and fps, and compute values for these properties for the monodepth video.

cap = cv2.VideoCapture(str(VIDEO_FILE))
ret, image =
if not ret:
    raise ValueError(f"The video at {VIDEO_FILE} cannot be read.")
input_fps = cap.get(cv2.CAP_PROP_FPS)
input_video_frame_height, input_video_frame_width = image.shape[:2]

target_fps = input_fps / ADVANCE_FRAMES
target_frame_height = int(input_video_frame_height * SCALE_OUTPUT)
target_frame_width = int(input_video_frame_width * SCALE_OUTPUT)

    f"The input video has a frame width of {input_video_frame_width}, "
    f"frame height of {input_video_frame_height} and runs at {input_fps:.2f} fps"
    "The monodepth video will be scaled with a factor "
    f"{SCALE_OUTPUT}, have width {target_frame_width}, "
    f" height {target_frame_height}, and run at {target_fps:.2f} fps"
The input video has a frame width of 640, frame height of 360 and runs at 30.00 fps
The monodepth video will be scaled with a factor 0.5, have width 320,  height 180, and run at 15.00 fps

Do Inference on a Video and Create Monodepth Video

# Initialize variables.
input_video_frame_nr = 0
start_time = time.perf_counter()
total_inference_duration = 0

# Open the input video
cap = cv2.VideoCapture(str(VIDEO_FILE))

# Create a result video.
out_video = cv2.VideoWriter(
    (target_frame_width * 2, target_frame_height),

num_frames = int(NUM_SECONDS * input_fps)
total_frames = cap.get(cv2.CAP_PROP_FRAME_COUNT) if num_frames == 0 else num_frames
progress_bar = ProgressBar(total=total_frames)

    while cap.isOpened():
        ret, image =
        if not ret:

        if input_video_frame_nr >= total_frames:

        # Only process every second frame.
        # Prepare a frame for inference.
        # Resize to the input shape for network.
        resized_image = cv2.resize(src=image, dsize=(network_image_height, network_image_width))
        # Reshape the image to network input shape NCHW.
        input_image = np.expand_dims(np.transpose(resized_image, (2, 0, 1)), 0)

        # Do inference.
        inference_start_time = time.perf_counter()
        result = compiled_model([input_image])[output_key]
        inference_stop_time = time.perf_counter()
        inference_duration = inference_stop_time - inference_start_time
        total_inference_duration += inference_duration

        if input_video_frame_nr % (10 * ADVANCE_FRAMES) == 0:
            # input_video_frame_nr // ADVANCE_FRAMES gives the number of
            # Frames that have been processed by the network.
                    f"Processed frame {input_video_frame_nr // ADVANCE_FRAMES}"
                    f"/{total_frames // ADVANCE_FRAMES}. "
                    f"Inference time per frame: {inference_duration:.2f} seconds "
                    f"({1/inference_duration:.2f} FPS)"

        # Transform the network result to a RGB image.
        result_frame = to_rgb(convert_result_to_image(result))
        # Resize the image and the result to a target frame shape.
        result_frame = cv2.resize(result_frame, (target_frame_width, target_frame_height))
        image = cv2.resize(image, (target_frame_width, target_frame_height))
        # Put the image and the result side by side.
        stacked_frame = np.hstack((image, result_frame))
        # Save a frame to the video.

        input_video_frame_nr = input_video_frame_nr + ADVANCE_FRAMES
        cap.set(1, input_video_frame_nr)

        progress_bar.progress = input_video_frame_nr

except KeyboardInterrupt:
    print("Processing interrupted.")
    processed_frames = num_frames // ADVANCE_FRAMES
    end_time = time.perf_counter()
    duration = end_time - start_time

        f"Processed {processed_frames} frames in {duration:.2f} seconds. "
        f"Total FPS (including video processing): {processed_frames/duration:.2f}."
        f"Inference FPS: {processed_frames/total_inference_duration:.2f} "
    print(f"Monodepth Video saved to '{str(result_video_path)}'.")
Processed 60 frames in 9.15 seconds. Total FPS (including video processing): 6.56.Inference FPS: 55.76
Monodepth Video saved to 'output/Coco Walking in Berkeley_monodepth.mp4'.

Display Monodepth Video

video = Video(result_video_path, width=800, embed=True)
if not result_video_path.exists():
    raise ValueError("OpenCV was unable to write the video file. Showing one video frame.")
    print(f"Showing monodepth video saved at\n{result_video_path.resolve()}")
        "If you cannot see the video in your browser, please click on the "
        "following link to download the video "
    video_link = FileLink(result_video_path)
    video_link.html_link_str = "<a href='%s' download>%s</a>"
Showing monodepth video saved at
/opt/home/k8sworker/cibuilds/ov-notebook/OVNotebookOps-408/.workspace/scm/ov-notebook/notebooks/201-vision-monodepth/output/Coco Walking in Berkeley_monodepth.mp4
If you cannot see the video in your browser, please click on the following link to download the video
output/Coco Walking in Berkeley_monodepth.mp4