Piero Toffanin pierotofy

## gist:4b98e26825ad866bd3801d8b9d1c9275
ffmpeg -y -i %d.png -filter_complex "tpad=stop_mode=clone:stop_duration=4,fps=14,scale=480:-1:flags=lanczos,split[s0][s1];[s0]palettegen=max_colors=32[p];[s1][p]paletteuse=dither=bayer" output.gif

## rolling-shutter-estimation.txt
1. Take a picture of a computer monitor with high shutter speed
2. Count number of lines in picture
3. If monitor frequency = F = 60 hz , number of lines = L, then readout time =~ (1 / F * 1000) * L (ms)

## launch-node.bat
cd resources\app\apps\NodeODM
nodeodm.exe --odm_path ..\ODM -p 3000

## plane.py
import numpy as np
import random
import math
import os

# https://stackoverflow.com/questions/38754668/plane-fitting-in-a-3d-point-cloud

def PCA(data, correlation = False, sort = True):
    """ Applies Principal Component Analysis to the data

## debug_plot.py
def debug_plot(data, common_tracks, pair, inliers, outliers):
    logger.info("Plotting %s" % str(pair))
    import matplotlib.pyplot as pl
    from opensfm import features
    pair_tracks, p1, p2 = common_tracks[pair]

    im1 = data.load_image(pair[0])
    im2 = data.load_image(pair[1])

    h1, w1, c = im1.shape

## Rtnd_from_H.py
def Rtnd_from_H(H, K, K1, reference_points=None, reference_normal=None):
    num_solutions, candidate_rotations, candidate_translations, candidate_normals = cv2.decomposeHomographyMat(H, K)

    if reference_points is not None:
        # Apply depth constraints to isolate two solutions
        # https://hal.inria.fr/inria-00174036/PDF/RR-6303.pdf
        # 3.3.2 reference-point visibility


        # Calculate projective coordinates of reference points

## gist:e028a3b90be0ebd12e1d7bd61a5352ae

      
              1 file
            
          
              0 forks
            
          
              0 comments
            
          
              0 stars
            
          
                pierotofy
                / gist:e028a3b90be0ebd12e1d7bd61a5352ae
            
            
              Created
              January 25, 2021 20:40
            
              
                Setup Minio Gateway on Windows
              
          
    Command Prompt #1

This simulates a AWS or Wasabi instance on localhost:9000
set MINIO_ROOT_USER=minioadmin
set MINIO_ROOT_PASSWORD=minioadmin
minio server ./data


## NodeCOLMAP
Goal: long focal lengths cameras

Colmap seem to work well with those.

Steps:
======

1. Add COLMAP, figure out command line pipeline, stages
2. Add dem2mesh
3. Add mvs-texturing

## gist:05dac5a81d5de14954648052b2ec30bd
exiftool -a -G1 -s -time:all name_0001.jpg

## export_colmap_mod.py
#!/usr/bin/env python3
# Copyright (c) 2018, ETH Zurich and UNC Chapel Hill.
# All rights reserved.
#
# Redistribution and use in source and binary forms, with or without
# modification, are permitted provided that the following conditions are met:
#
#     * Redistributions of source code must retain the above copyright
#       notice, this list of conditions and the following disclaimer.
#
	1. Take a picture of a computer monitor with high shutter speed
	2. Count number of lines in picture
	3. If monitor frequency = F = 60 hz , number of lines = L, then readout time =~ (1 / F * 1000) * L (ms)
	cd resources\app\apps\NodeODM
	nodeodm.exe --odm_path ..\ODM -p 3000
	import numpy as np
	import random
	import math
	import os

	# https://stackoverflow.com/questions/38754668/plane-fitting-in-a-3d-point-cloud

	def PCA(data, correlation = False, sort = True):
	""" Applies Principal Component Analysis to the data
	def debug_plot(data, common_tracks, pair, inliers, outliers):
	logger.info("Plotting %s" % str(pair))
	import matplotlib.pyplot as pl
	from opensfm import features
	pair_tracks, p1, p2 = common_tracks[pair]

	im1 = data.load_image(pair[0])
	im2 = data.load_image(pair[1])

	h1, w1, c = im1.shape
	def Rtnd_from_H(H, K, K1, reference_points=None, reference_normal=None):
	num_solutions, candidate_rotations, candidate_translations, candidate_normals = cv2.decomposeHomographyMat(H, K)

	if reference_points is not None:
	# Apply depth constraints to isolate two solutions
	# https://hal.inria.fr/inria-00174036/PDF/RR-6303.pdf
	# 3.3.2 reference-point visibility


	# Calculate projective coordinates of reference points
	Goal: long focal lengths cameras

	Colmap seem to work well with those.

	Steps:
	======

	1. Add COLMAP, figure out command line pipeline, stages
	2. Add dem2mesh
	3. Add mvs-texturing
	#!/usr/bin/env python3
	# Copyright (c) 2018, ETH Zurich and UNC Chapel Hill.
	# All rights reserved.
	#
	# Redistribution and use in source and binary forms, with or without
	# modification, are permitted provided that the following conditions are met:
	#
	# * Redistributions of source code must retain the above copyright
	# notice, this list of conditions and the following disclaimer.
	#