In DavidDHofmann/video2pic: What the Package Does (One Line, Title Case)

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.path = "man/figures/README-",
  out.width = "100%"
)

video2pic

video2pic is an R-package that allows you to extract still frames from a video. For efficiency purposes, the frame-extraction-algorithm has been implemented in python.

Installation

Because the algorithm uses python in the background, you first need to install miniconda on your system. Please follow the instructions here to do so. Once you have installed miniconda, you can install the development version of video2pic from GitHub with:

# install.packages("devtools")
devtools::install_github("DavidDHofmann/video2pic")

Finally, you will also need to ensure that all necessary python packages are installed. This can be done using the video2pic_install() function.

library(video2pic)
video2pic_install()

This will create a virtual python environment called "video2pic" containing the required packages. You may need to restart your R-session after this!

Example

The function to extract frames from a video is called video2pic(). The function allows you to specify an output directory and the number of frames per second that should be extracted. Note that you could also provide multiple videos at once, so that you don't need to loop through multiple files.

# Load required packages
library(video2pic)
library(magick)

# Initalize python environment
video2pic_initialize()

# Path to the example video
video <- system.file("extdata", "Earth.mp4", package = "video2pic")

# Extract one frame per second (and store them to a temporary directory)
video2pic(video, outdir = tempdir(), fps = 1)

# The extracted images can now be found here
imgs <- dir(tempdir(), pattern = ".JPG", full.names = T)

# We can visualize some of them if we want
image_read(sample(imgs, 4))