WindowsでOpenMVGを自動実行するためのPythonスクリプト

Pythonスクリプトはビルド先フォルダ以下のsoftware\SfMフォルダにSfM_SequentialPipeline.pyなどで生成されています。IncrementalSfMを実行する場合は、このSfM_SequentialPipeline.pyを使います。主に下記で赤字の部分を環境に合わせて変更して、第一引数に写真を格納したフォルダ名、第二引数に出力先フォルダ名を指定します。実行すると出力先フォルダ以下のmatchesフォルダに特徴点情報のファイル、reconstruction_sequentialフォルダに再構築結果が出力されます。

C:/Git/openMVG/src/openMVG/exif/sensor_width_database/sensor_width_camera_database.txtがセンサーサイズを定義したデータベースファイルです。定義方法などはOpenMVGをVisual Studio 2017(x64)でビルド(その2)を参照してください。

先頭行の"＃ coding: Shift_JIS"は、ファイルパスに日本語が入るときのために定義しています。

 

# coding: Shift_JIS

# Python implementation of the bash script written by Romuald Perrot
# Created by @vins31
# Modified by Pierre Moulon
#
# this script is for easy use of OpenMVG
#
# usage : python openmvg.py image_dir output_dir
#
# image_dir is the input directory where images are located
# output_dir is where the project must be saved
#
# if output_dir is not present script will create it
#

# Indicate the openMVG binary directory
OPENMVG_SFM_BIN = "C:/Git/openMVG/build/Windows-AMD64-Release/Release"

# Indicate the openMVG camera sensor width directory
CAMERA_SENSOR_WIDTH_DIRECTORY = "C:/Git/openMVG/src/openMVG/exif/sensor_width_database"

#import commands
import os
import subprocess
import sys

if len(sys.argv)
    print ("Usage %s image_dir output_dir" % sys.argv[0])
    sys.exit(1)

input_dir = sys.argv[1]
output_dir = sys.argv[2]
matches_dir = os.path.join(output_dir, "matches")
reconstruction_dir = os.path.join(output_dir, "reconstruction_sequential")
camera_file_params = os.path.join(CAMERA_SENSOR_WIDTH_DIRECTORY, "sensor_width_camera_database.txt")

print ("Using input dir  : ", input_dir)
print ("      output_dir : ", output_dir)

# Create the ouput/matches folder if not present
if not os.path.exists(output_dir):
  os.mkdir(output_dir)
if not os.path.exists(matches_dir):
  os.mkdir(matches_dir)

print ("1. Intrinsics analysis")
pIntrisics = subprocess.Popen( [os.path.join(OPENMVG_SFM_BIN, "openMVG_main_SfMInit_ImageListing"),  "-i", input_dir, "-o", matches_dir, "-d", camera_file_params, "-c", "4"] )
pIntrisics.wait()

print ("2. Compute features")
pFeatures = subprocess.Popen( [os.path.join(OPENMVG_SFM_BIN, "openMVG_main_ComputeFeatures"),  "-i", matches_dir+"/sfm_data.json", "-o", matches_dir, "-m", "SIFT"] )
pFeatures.wait()

print ("3. Compute matches")
pMatches = subprocess.Popen( [os.path.join(OPENMVG_SFM_BIN, "openMVG_main_ComputeMatches"),  "-i", matches_dir+"/sfm_data.json", "-o", matches_dir] )
pMatches.wait()

# Create the reconstruction if not present
if not os.path.exists(reconstruction_dir):
    os.mkdir(reconstruction_dir)

print ("4. Do Sequential/Incremental reconstruction")
pRecons = subprocess.Popen( [os.path.join(OPENMVG_SFM_BIN, "openMVG_main_IncrementalSfM"),  "-i", matches_dir+"/sfm_data.json", "-m", matches_dir, "-o", reconstruction_dir, "-c", "4"] )
pRecons.wait()

print ("5. Colorize Structure")
pRecons = subprocess.Popen( [os.path.join(OPENMVG_SFM_BIN, "openMVG_main_ComputeSfM_DataColor"),  "-i", reconstruction_dir+"/sfm_data.bin", "-o", os.path.join(reconstruction_dir,"colorized.ply")] )
pRecons.wait()

# optional, compute final valid structure from the known camera poses
print ("6. Structure from Known Poses (robust triangulation)")
pRecons = subprocess.Popen( [os.path.join(OPENMVG_SFM_BIN, "openMVG_main_ComputeStructureFromKnownPoses"),  "-i", reconstruction_dir+"/sfm_data.bin", "-m", matches_dir, "-f", os.path.join(matches_dir, "matches.f.bin"), "-o", os.path.join(reconstruction_dir,"robust.bin")] )
pRecons.wait()

pRecons = subprocess.Popen( [os.path.join(OPENMVG_SFM_BIN, "openMVG_main_ComputeSfM_DataColor"),  "-i", reconstruction_dir+"/robust.bin", "-o", os.path.join(reconstruction_dir,"robust_colorized.ply")] )
pRecons.wait()