Inspect: Component Connectivity vs Match Count

In the Inspect tool, what is the difference between Component Connectivity and Match Count?

They seem to have the same/similar definitions. They seem to both define whether 2 cameras are part of the same virtual component.

Component Connectivity

If there is at least this number of edges among cameras, then those cameras are connected into one component (cameras marked with the same color).

Match Count

Two cameras are connected into one virtual component only if there is at least this number of common features (matches) among those two cameras

I understand Match Count better. Matches are defined by the Apical Angle and Feature Consistency parameters. Match Count is count of common matches between 2 cameras.

But Component Connectivity is more vaguely defined.

Hi Will,
component connectivity is more general and it is related to the cameras. This settings set the minimal number how many camera’s edges are creating one virtual component.
Like these cameras have three edges and create one pink virtual component:

And as you wrote, match count is the feature count between two cameras.

Ok so for Component Connectivity, does it mean:

number of edges among cameras

This is the total number of edges that exist among a group of cameras

those cameras are connected into one component (cameras marked with the same color)

Ok so if we set this to “4”, then you can’t have a component with only 3 cameras (max number of edges being 3). But you could have a component with 4 or more cameras (max edges for 4 cameras is 6).

So for cameras below this threshold they just stay by themselves in a 1-camera virtual component, and cameras above this threshold end up in a multi-camera virtual-component.

Is that what’s going on, @OndrejTrhan ?

it’s basically a way to keep smaller virtual-components from grouping up.

I suppose it should be like that.