Six Colors
Six Colors

by Jason Snell & Dan Moren

This Week's Sponsor

End users aren't your enemy! Kolide gets users to fix their own device compliance problems–and unsecure devices can't log in. Click here to learn how.

By Jason Snell

Creating podcast video with a 360-degree camera

The view from the top.

Recently, the podcasters of The Incomparable got together in person and one of the things we did was exactly what you’d expect podcasters to do—we made podcasts. Given the unique situation of us being in the same room together, I decided to see if I could capture the podcasts on video, too.

The challenge is that I didn’t want to set up a bunch of cameras. I didn’t want the addition of video to be too invasive. For years now, I’ve imagined the video version of a podcast happening around a table as looking like the circle scene from “That ’70s Show.”1 A single camera at the center of the table can shoot the entire conversation.2

Fortunately, these days you can actually buy off-the-shelf 360-degree cameras. They’re mostly made for action sports, and they have some pretty spectacular effects—there’s enough overlap between the two 180-degree lenses on both sides of the camera that the tripod or selfie stick mount disappears, leaving the impression that the video is coming from a tiny drone hovering very close to the action.

The Insta360 X3 camera sat just in front of our table and captured the whole scene. (Photo: Chip Sudderth.)

I tried two different 360-degree cameras, the $450 Insta360 X3 and the $500 GoPro Max. Both capture 360 video at 5.6K resolution (5760×2880) —which is good, because it takes a lot of cropping to get usable shots out of that file. Both of the cameras were a bit janky—they feel a bit underpowered, as if this category of product is only barely viable.

In the end, I returned the Insta360 X3 and kept the GoPro Max for a single reason: Every 30 minutes, the Insta360 would stop recording, pause for 10 seconds, and then start recording again. That’s not great for multi-hour podcast sessions. The GoPro Max made a new video file every eight minutes, but it didn’t drop frames and would keep recording until its SD card was full.

To run these cameras for the length of time of a podcast, their batteries just won’t do. Instead, you have to run them with their battery door open and a USB cable plugged in. It seems like this helps with cooling a little bit too, which is good because these things get pretty hot when they’re capturing video. The GoPro comes with lens caps and special transparent lens protectors that somewhat degrade quality but will protect from crashes if you’re doing to take it snowboarding or mountain biking. Podcasts are less dangerous, usually.

Capturing the video was relatively easy. Once I had plugged the camera into power, I placed it in the middle of the action using the Insta360 Selfie Stick with Tripod, which was so good—it’s thin, light, and extensible—that I kept it even after I switched to the GoPro.

Getting the video off the cameras afterward is a little tricky. There’s a lot of magic going on in a 360 capture—it appears that the cameras capture two 180-degree fields as well as a bunch of position data (again, more relevant when you’re skiing than podcasting). To get a usable video file out, you need to use their smartphone or desktop apps.

The desktop apps are bad, but they did the job. The Insta360 and GoPro apps both let you set up crops and pans and then export the result, which would be perfect for the quick-and-dirty sharing of action footage that these cameras are designed for. For podcasting, I decided to try two different approaches: exporting individual crops one by one, or exporting a full 360-degree video file and dealing with cropping and panning in Final Cut Pro later.

Pan down to the D&D map.

I rapidly realized that framing individual shots and exporting them one by one was going to take too much time and consume too much disk space, especially when I discovered that Final Cut Pro innately understands 360 video and provides tools to zoom, pan, tilt, and roll to get the right shots for a traditional 16:9 video. Not only that, but working with the raw 360 video means you can change your shots whenever you want, and even use keyframe animation to create virtual camera moves, all out of a single video file.

Final Cut’s Orientation panel gives you control over the field of view in the 360 shot.

My ultimate goal was to create multiple shots from the 360 video, focused on one or two people, to make it look like my podcast was shot with multiple cameras. Unfortunately, Final Cut Pro doesn’t let you define different orientations in a single 360 clip as virtual cameras, so I wasn’t able to use the app’s multi-camera switcher unless I went back to laboriously exporting every shot in advance and then importing those clips as multi-cam shots.

My shots as video effects presets.

Instead, I found a workaround: It turns out that you can save any collection of video settings, including 360 pan, zoom, and field of view, as a Video Effects Preset file, which shows up in Final Cut Pro’s Effects bin. Now I can set up each shot for a project, save it as a Preset, and then just apply those “effects” to each clip to simulate switching between cameras.

But applying effects in Final Cut Pro requires that you drag and drop an effect icon out of the bin and onto the clip, or alternatively that you select a clip and then double-click on an effect in the bin. Amazingly, in an app that’s got keyboard shortcuts for just about everything, there doesn’t seem to be any way to assign specific effects to specific keystrokes.

Not being able to quickly assign “cameras” to each shot was going to make editing these sessions a huge pain, so I brought out the big guns, namely Keyboard Maestro’s screen-scraping feature. I built a Keyboard Maestro macro that looks for the Effects panel in the Final Cut Pro interface and then double clicks at an offset point that aligns perfectly with the topmost saved Preset (and then moves the pointer back where it was). Then I duplicated that macro and changed the offset point to align with the second Preset, and so on. Then I assigned all those Keyboard Maestro macros to buttons on my Stream Deck.

Now my workflow is pretty straightforward: I can click on a clip in the timeline, press a Button on my Stream Deck to select which “camera” I’m using, and the clip will change to the assigned shot. It works, but really, Final Cut Pro should allow you to assign keystrokes to effects. Even better, it should allow users to define “cameras” in 360 footage and use them as a multicam clip.

In any event, I’m happy with my experimentation with using a 360 camera to capture podcast video. The video quality is acceptable—we were dealing with some pretty spotty lighting conditions—and editing the result in Final Cut has proven to be a manageable, if quirky, process.

  1. No drugs stronger than caffeine were consumed during these podcasts. 
  2. I know there are some videoconferencing solutions that will automatically focus on who is speaking around a table, but they didn’t really mesh with what I wanted to do. 

If you appreciate articles like this one, support us by becoming a Six Colors subscriber. Subscribers get access to an exclusive podcast, members-only stories, and a special community.

Search Six Colors