Weapons & threatsCCTV Knife Detection
Synthetic knife detection dataset: 1,000 CCTV images of people holding knives plus look-alike negatives, labelled person and knife in YOLO and COCO. Free sample.
- Size
- 1,000 images
- Format
- YOLO + COCO
Weapons
A person turns and walks past a shop door at night with a large kitchen knife held low in the right hand.
This clip uses the same camera and the same street as the handgun clip. The camera is fixed high on the corner of a shop and looks down a lit pavement at night, with parked cars and a street beyond. Because the view is identical, the only thing that changes between the two clips is the object in the person's hand.
The person wears a dark jacket, a cap and grey trousers. A large kitchen knife is held low in the right hand, down by the hip. It is clearly visible from about the midpoint of the clip, when the person turns and walks past the shop door. Before that the knife is mostly hidden by the body and the angle, which is the situation a knife detector has to handle: a thin object, held low, against a busy background.
No detector overlay is drawn on this clip. You see the scene as a camera would record it, without boxes or scores. That makes it useful for judging how visible the knife is to a person watching the footage, and how small it is compared with the rest of the frame.
The knife was generated into the empty scene, and the original footage without it is shown lower on this page. The dataset for this scenario contains person and knife labels in YOLO and COCO formats, and a free sample can be downloaded to check the labels against your own detector.
The generated clip above uses this same camera and scene.
Dataset
Weapons & threatsSynthetic knife detection dataset: 1,000 CCTV images of people holding knives plus look-alike negatives, labelled person and knife in YOLO and COCO. Free sample.
More clips

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.

From a moving drone, three civilians walk a dirt track and a group of armed figures enters from the left.

From a moving drone, three civilians walk a track and near the end a main battle tank drives in beside the houses.
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.