Custom video
Hospital corridor threat clip from one empty frame
A man in a checked shirt steps out of a door in an empty hospital corridor and walks away carrying a rifle.
What the clip shows
The camera is a wide fisheye mounted on the ceiling of a hospital corridor. The picture has a green tint, as many cameras do in night mode. Doors line both sides of the corridor and trolleys stand against the wall. For the first moments the corridor is empty, which is what this camera sees for most of the day.
Then a man in a checked shirt steps out of a door on the left. He is carrying a rifle, and he walks down the corridor towards the far end. The fisheye lens bends the walls and makes the figure shrink quickly as he moves away, so the weapon is large when he appears and small a second later.
This clip was generated from a single frame of the empty corridor. That frame is the customer's own camera view, and the event was generated into the empty scene. It is an example of a custom video dataset: the view, the angle and the lens distortion are the ones the customer's detector will meet in production, instead of a generic stock scene.
It is not a clip from a catalogue dataset, and no catalogue dataset page describes it. Custom video clips are quoted per project, and can come with boxes, tracking IDs and event times. The catalogue weapon detection dataset is the image-based starting point if you want labelled rifles and handguns without a custom project.
- Camera
- Wide fisheye ceiling camera, green night-mode tint
- Scene
- Hospital corridor with doors on both sides and trolleys against the wall
- Event
- Man in a checked shirt steps out of a left-hand door with a rifle and walks to the far end
- Source
- One frame of the empty corridor, the customer's own camera view
- Clip length
- 6 seconds, 24 fps
- Resolution
- 560 x 560
Dataset
Made on your own cameras
This clip is not part of a catalogue dataset. It shows what a custom video dataset looks like: the event is generated into an empty frame of your own camera view, so the angle, lens and lighting match your deployment.
Custom video datasetsMore clips
Other sample videos

CCTV handgun detection: a labelled night-time clip
A person walks along a lit shop front at night with a handgun in the right hand, followed frame by frame by a red weapon box.

CCTV knife detection: a kitchen knife on a night street
A person turns and walks past a shop door at night with a large kitchen knife held low in the right hand.

Drone detection of armed people among civilians
From a moving drone, three civilians walk a dirt track and a group of armed figures enters from the left.
Need this event on your own cameras?
Send 5–20 photos of your empty scenes and we render the event into them, as images or video clips with boxes, tracking IDs and event times.