This is primarily a note to self post, but could hopefully also be useful for others. At least, I spent a little too long to figure out to embed a YouTube video with a specific language on the subtitles.
The starting point is that I had this project video that I wanted to embed on a project website:
However, then I found that you can add info about the specific language you want to use by adding this snippet after the URL:
This means ?hl=en is the language of the controls, &cc_lang_pref=en is the language of the subtitles and &cc=1 turns on the subtitles. The complete block is:
Continuing my explorations of FFmpeg for video visualization, today I came across this very nice blog post on creating “pixel array” images of videos. Here the idea is to reduce every single frame into only one pixel, and to plot this next to each other on a line. Of course, I wanted to try this out myself.
I find that creating motiongrams or videograms is a good way to visualize the content of videos. They are abstract representations, but still reveal some of what is going on. However, for longer videos, motiongrams may be a bit tricky to look at, and they also take a lot of time to generate (hours, or even days). For that reason I was excited to see how pixel array images would work on some of my material.
First I tried on my “standard” dance video:
which gives this pixel array image:
Yes, that is mainly a blue line, resulting from the average colour of the video being blue throughout the entire video.
And, yes, that is mainly a gray line… I realize that this method does not work very well with single-shot videos.
To try something very different, I also decided to make a pixel array image of Bergensbanen, a 7-hour TV production of the train between Oslo and Bergen. I made videograms of this recording some years ago, which turned out to be quite nice. So I was excited to see how a pixel array image would work. The end result looks like this (1920 pixels wide):
As you see, not much is changing, but that also represents the slowness of the train ride. While I originally thought this would be a smart representation, I still think that my videograms were more informative, such as this one:
The big difference between the two visualizations, is that each frame is represented with vertical information in the videogram. The pixel array image, on the other hand, only displays one single pixel per frame. That said, it took only some minutes to generate the pixel array image, and I recall spending several days on generating the videogram.
To sum up, I think that pixel array images are probably more useful for movies and video material in which there are lots of changes throughout. They would be better suited for such a reduction technique. For my videos, in which I always use single-shot stationary cameras, motiongrams and videograms may still be the preferred solution.
This is a note to self, and could potentially also be useful to others in need of converting “old-school” MPEG-2 files into more modern MPEG-4 files using FFmpeg.
In the fourMs lab we have a bunch of Canon XF105 video cameras that record .MXF files with MPEG-2 compression. This is not a very useful format for other things we are doing, so I often have to recompress them to something else.
Inspecting one of the files, I just also discovered that they record the audio onto two mono channels:
So I also want to merge these two mono tracks (which are the left and right inputs of the camera) to a stereo track. FFmpeg comes in handy (as always), and I figured out that this little one-liner will do the trick:
Very many people are currently moving to video-based meetings. For that reason I have written up some quick advise on how to improve your setup. This is based on my interview advise, but grouped differently.
The first important thing is to have as good a network as you can. Video conferencing requires a lot of bandwidth, so even though your e-mail and regular browsing works fine, it may still not be sufficient for good video transmission.
Cabled network: If you are able to connect with an Ethernet cable to your router, that would usually always be the best and most solid solution.
Wireless network: If cable won’t work for you (it is also difficult logistically in my own apartment), try to get as close as possible to your wi-fi router.
I would argue that improving the audio is more important than the video for video conferencing. Most video conferencing systems (Skype, Zoom, etc.) will prioritize the audio channel, which means that the video may stutter while the audio is passing through fine.
The main trick is to aim for separating the “foreground” as much as possible from the “background”. There are some very basic audio principles to follow:
Use a headset: The best way to get decent sound for video conferencing, is to move the microphone as close as possible to your mouth. Headsets with a microphone boom in front of your face are the best, but a regular mobile phone headset (the one that came with your mobile phone, for example) would still be better than nothing.
Use headphones: If you for some reason do not have a headset with built-in microphone, using a regular pair of headphones is still better than using the speakers on your computer. With this setup you use the microphone on the computer, which may not be ideal, but at least you won’t get feedback problems.
Avoid reverberant rooms: If you aim for clarity in conversation, it is typically better to sit in a smaller and more damped room than a large one. That means that a bedroom is typically better than a larger living room. If you use a headset this is less important, but particularly if you only use the built-in microphone and speakers on a laptop, this could make a huge difference in how your voice gets through.
Mute yourself: In most system there is a button to mute yourself. If you are not talking all the time, it helps to mute yourself from the discussion. Just remember to unmute when you want to say something!
The same principle of separating “foreground” from “background” applies to the video.
Lighting: To obtain the best possible video image, think about your placement with respect to lighting. It is, for example, not ideal to sit in front of a window, since a bright light in the background will make it difficult to see your face.
Background: The best is to sit in front of a plain wall. If that is not possible, consider whether the background of your image is what you want to show to your fellow students/colleagues.
Video angle: If you are using the built-in camera on your computer you may not have too many options for how to place the camera. But you may still consider shifting the camera position so that you and your surroundings look as good as possible.
There are, of course, many ways to improve your video conferencing setup. Many people believe that you need to invest in expensive equipment to get good results. But even cheap consumer products are very capable of producing decent results these days. So it is more a matter of optimizing what you have. Good luck!
I am continuing my explorations of the great terminal-based video tool FFmpeg. Now I wanted to see if I could “flatten” a 360-degree video recorded with a Ricoh Theta camera. These cameras contain two fisheye lenses, capturing two 180-degree videos next to each other. This results in video files like shown in the screenshot below.
These files are not very useful to watch or work with, so we need to somehow “flatten” it into a more meaningful video file. I find it cumbersome to do this in the Ricoh mobile phone apps, so have been looking for a simple solution to do it on my computer.
I see that the FFmpeg developers are working on native support for various 360-degree video files. This is implemented in the filter v360, but since it is not in the stable version of FFmpeg yet, I decided to look for something that works right now. Then I came across this blog post, which shows how to do the flattening based on two so-called PGM files that contain information about how the video should be mapped: