Skip to content
NAMEFRAMEPricing
Pricing

Large objects on flat concrete, in hard light.

The reference capture, and the easiest of the nine: big objects, high contrast, a background that holds still. It is the scene to read first, because when a pipeline gets a label wrong here it is the pipeline that is wrong and not the picture.

Camera
Four authored zones looking at a target, 1280×960
Frames in the run
300
Labelled instances
32,546
Classes
6
Validator grade
A · 92.1/100
Airbase: the hangar apron: hard shadows, vehicles and scattered ground clutter
Airbase: the hangar apron: hard shadows, vehicles and scattered ground clutterCaptured with Unreal Engine 5.8

What this scene is hard for

The reason it is one of the nine, rather than a prettier map that would have proved less.

  • Hard shadow on flat concrete, which has the size and the edges of an object and is not one.
  • Repeated identical props — containers, barrels, crates in rows — that a detector will merge into one box given the chance.
  • Large objects and small ones in the same frame, from a tank to a person beside it.
  • Very little else. That is the point: the failures this scene produces are the pipeline's own.

Frames from the run

Chosen by eye out of the frames published for this capture, with what each one shows. They are the run's own contact-sheet images at the size the site publishes them, not stills re-exported for this page.

These six are close together in the run: the published strip is twelve consecutive frames of one camera zone, so it shows the apron from six angles rather than the airbase across a day. The scene has zones and hours the strip does not reach; the complete capture is on the dataset page.

  • The apron with the service road curving out of frame, and clutter spread the full depth of the picture
    The apron with the service road curving out of frame, and clutter spread the full depth of the picture
  • Lowest over the hangars, where the roofs fill the frame and the ground is only visible between them
    Lowest over the hangars, where the roofs fill the frame and the ground is only visible between them
  • Wide to the tree line, with the hardstanding and its vehicles across the middle distance
    Wide to the tree line, with the hardstanding and its vehicles across the middle distance
  • The establishing angle across the whole apron, which is the view the capture returns to
    The establishing angle across the whole apron, which is the view the capture returns to
  • The first frame of the run, lowest over the curved hangar with the scrub at the edge of the apron
    The first frame of the run, lowest over the curved hangar with the scrub at the edge of the apron
  • Across the apron, where the same white props repeat in a row and a detector is invited to merge them
    Across the apron, where the same white props repeat in a row and a detector is invited to merge them

Where this map came from

We did not model it. It is a third-party Unreal environment, dressed with a capture contract, and what NameFrame produced is the frames and the labels — no map and no .uasset is redistributed in any archive on this site. No author is named here because we have no agreement with one; if this scene is yours and you would like it credited, write to hello@getnameframe.com.

The data itself is on the capture’s page

This page is about the place. The archives, the validator’s findings, the class list, the checksums and what is wrong with the run are on the Airbase capture, which is the one page on this site that means this dataset.