Counter-UAS & defenceUAV & Aerial View: Battle Tank Detection
Synthetic aerial dataset for battle tank detection from UAV and drone perspectives. 3,000 YOLO-annotated images with 164 open-source samples on Kaggle.
- Size
- 3,000 images
- Format
- YOLO
Dataset library
16 synthetic computer vision datasets for CCTV and security.
16 synthetic datasets with YOLO and COCO labels, each with a free sample. Licences from $149.
Counter-UAS & defence
Behaviour & safetyCCTV Shoplifting Detection
Train a shoplifting detection model: 5,000+ synthetic CCTV frames and 100+ videos with YOLO, pose and VLM labels. Free 400-image sample on Kaggle.
- Size
- 5,000+ frames + 100+ videos
- Format
- YOLO + VLM
Weapons & threatsCCTV ATM Robbery Detection: Gun & Knife
Synthetic ATM robbery detection dataset with 3,000 YOLO images. Four classes: offender, victim, gun, knife. 237 open-source samples on Kaggle.
- Size
- 3,000 images
- Format
- YOLO
Fire & smokeCCTV Smoke & Fire Emergency Detection
Smoke and fire detection dataset: 1,000 synthetic CCTV images of early-stage fires plus look-alike negatives, person/smoke/fire boxes in YOLO and COCO.
- Size
- 1,000 images
- Format
- YOLO + COCO
Fire & smokeLong-Distance Wildfire & Smoke Detection
Wildfire smoke detection dataset: 1,000 synthetic fire-lookout images with smoke and wildfire boxes in YOLO and COCO, night fires and look-alike negatives.
- Size
- 1,000 images
- Format
- YOLO + COCO
Weapons & threatsCCTV Weapon Detection
Train AI weapon detection on 1,000 synthetic CCTV images: handguns, rifles, shotguns and look-alike negatives, labelled in YOLO and COCO. Free sample included.
- Size
- 1,000 images
- Format
- YOLO + COCO
Weapons & threatsCCTV Knife Detection
Synthetic knife detection dataset: 1,000 CCTV images of people holding knives plus look-alike negatives, labelled in YOLO and COCO. Free sample.
- Size
- 1,000 images
- Format
- YOLO + COCO
Behaviour & safetyCCTV Fall & Incident Detection
Synthetic fall detection dataset: 1,000 CCTV images of fallen people plus sitting and kneeling look-alikes, labelled person / fallen_person in YOLO and COCO.
- Size
- 1,000 images
- Format
- YOLO + COCO
Counter-UAS & defenceAirport Drone Threat & Safety
Synthetic airport drone detection dataset: 1,200 YOLO images, small-object focus, adverse weather, bird-vs-drone labels. GDPR compliant.
- Size
- 1,200 images
- Format
- YOLO
Behaviour & safetyCCTV Fight & Aggression Detection
Fight detection dataset: 1,000 synthetic CCTV images with person boxes, 17-keypoint poses and fight boxes in YOLO-pose and COCO, plus look-alike negatives.
- Size
- 1,000 images
- Format
- YOLO-pose + COCO
Behaviour & safetyPedestrian Attribute Recognition
Synthetic pedestrian attribute dataset — a PA-100k alternative with captions and structured labels for ViT, CLIP and Re-ID. Free 996-image sample.
- Size
- 10,000 images (50K by quote)
- Format
- JSONL (Descriptive + Structured)
Weapons & threatsCCTV Weapon Detection: Rifles vs Umbrellas
Synthetic hard-negative dataset: rifles vs umbrellas to cut weapon-detection false positives. 1,000 YOLO images. Open-source sample available.
- Size
- 1,000 images
- Format
- YOLO
Weapons & threatsWeapon Detection: Handgun vs Bag of Chips
Fix the 'Doritos Effect' in weapon detection AI with handgun vs chip-bag hard negatives. 1,100 YOLO images. Open-source sample available.
- Size
- 1,100 images
- Format
- YOLO
Counter-UAS & defenceMilitary Drone Swarm & Saturation Attack
Synthetic drone swarm dataset for Counter-UAS training. 5-15 drones per frame, 10-50 pixel objects, adverse weather. 1,000+ YOLO-annotated images.
- Size
- 1,000+ images
- Format
- YOLO
Counter-UAS & defenceDrone Civilian / Armed Military Detection
Synthetic drone-perspective dataset distinguishing civilians from armed personnel. 800 YOLO images for aerial surveillance AI training.
- Size
- 800 images
- Format
- YOLO
Access nowSample on request
Behaviour & safetyDriver Drowsiness Detection
Synthetic driver drowsiness dataset with 10,000 images and pixel-perfect eye keypoint annotations. 168 open-source samples. Ideal for fatigue detection AI.
- Size
- 10,000 images
- Format
- YOLO + Keypoints
This is the full list of Simuletic's computer vision datasets: 16 synthetic CCTV, drone and dashcam sets for security and safety detection, each sold as a one-time licence with YOLO and COCO labels.
Open a card for the full specification and licence. Free sample opens the Kaggle or Hugging Face download for that dataset, so you can check the labels before you buy.
What the catalogue covers
Four domains, sixteen datasets.
Weapons and threats
Handguns, rifles and knives seen from real CCTV heights, in crowds and bad light, plus the hard negatives — umbrellas, phones, chip bags — that cause most false alarms. ATM robbery data separates offender from victim.
- CCTV ATM Robbery Detection: Gun & Knife · 3,000 images
- CCTV Weapon Detection · 1,000 images
- CCTV Knife Detection · 1,000 images
- CCTV Weapon Detection: Rifles vs Umbrellas · 1,000 images
- Weapon Detection: Handgun vs Bag of Chips · 1,100 images
Behaviour and safety
Shoplifting and concealment, fights and aggression with pose keypoints, falls and collapses, pedestrian attributes for search and re-identification, and in-cabin driver drowsiness — with keypoints, captions or video where the task needs them.
- CCTV Shoplifting Detection · 5,000+ frames + 100+ videos
- CCTV Fall & Incident Detection · 1,000 images
- CCTV Fight & Aggression Detection · 1,000 images
- Pedestrian Attribute Recognition · 10,000 images (50K by quote)
- Driver Drowsiness Detection · 10,000 images
Fire and smoke
Early-stage fires and smoke from street and site cameras, and long-range wildfire plumes from lookout towers, where the target can be a few pixels wide.
- CCTV Smoke & Fire Emergency Detection · 1,000 images
- Long-Distance Wildfire & Smoke Detection · 1,000 images
Counter-UAS and defence
Drones against airport backgrounds, swarms of small drones in clutter, armour from UAV altitude, and civilian-versus-armed classification — data that is restricted or dangerous to collect for real.
- UAV & Aerial View: Battle Tank Detection · 3,000 images
- Airport Drone Threat & Safety · 1,200 images
- Military Drone Swarm & Saturation Attack · 1,000+ images
- Drone Civilian / Armed Military Detection · 800 images
Free samples
Free open-source samples on Kaggle and Hugging Face.
Almost every dataset has a free open-source sample — typically 100 to 400 labelled images, plus video clips where the dataset has them. Train a baseline, look at the boxes, and check the camera angles against your own before you buy.
Samples are published on Kaggle and, for some datasets such as weapon and knife detection, on Hugging Face. Each dataset page links its sample. The samples carry their own open licence; the full datasets are licensed separately.
- UAV & Aerial View: Battle Tank Detection164 free
- CCTV Shoplifting Detection400 free
- CCTV ATM Robbery Detection: Gun & Knife237 free
- CCTV Smoke & Fire Emergency Detection240 free
- Long-Distance Wildfire & Smoke Detection239 free
- CCTV Weapon Detection141 free
- CCTV Knife Detection114 free
- CCTV Fall & Incident Detection111 free
Licences
How licensing works.
Per dataset, one-time
Every licence is a one-time payment for 1,000 images of one dataset, with no subscription. Research costs $149 for students and non-commercial work; commercial costs $590. Packs of 5,000–20,000 images are commercial only, $2,400–$8,500; video datasets are priced on request.
What's in the files
Full-resolution images, YOLO and COCO labels, a fixed train / val split and a dataset card describing classes, scenes and how the labels were made. Some datasets add keypoints, VLM captions or video sequences — the spec box on each page lists them.
Sample first, then buy
Download the free sample, train a baseline and check the labels. Then buy the licence from your account; downloads stay there. Prices and terms are on the pricing page.
Compare
All 16 datasets side by side.
| Dataset | Images | Classes | Hard negatives | Video | Free sample | Format | Licence |
|---|---|---|---|---|---|---|---|
| UAV & Aerial View: Battle Tank Detection | 3,000 images | battle_tank | — | No | 164 images | YOLO | $149 research / $590 commercial |
| CCTV Shoplifting Detection | 5,000+ frames + 100+ videos | person, pose-17kp | — | Video sequences | 400 images | YOLO + VLM | Price on request |
| CCTV ATM Robbery Detection: Gun & Knife | 3,000 images | offender, victim, gun, knife | — | No | 237 images | YOLO | $149 research / $590 commercial |
| CCTV Smoke & Fire Emergency Detection | 1,000 images | person, smoke, fire | ~16%: steam, fog, dust, lights, reflections | No | 240 images | YOLO + COCO | $149 research / $590 commercial |
| Long-Distance Wildfire & Smoke Detection | 1,000 images | smoke, wildfire | ~28%: cloud, fog, mist, dust, steam | No | 239 images | YOLO + COCO | $149 research / $590 commercial |
| CCTV Weapon Detection | 1,000 images | person, handgun, rifle, shotgun, sniper_rifle | ~20%: bottles, umbrellas, bags, phones, tools | No | 141 images | YOLO + COCO | $149 research / $590 commercial |
| CCTV Knife Detection | 1,000 images | person, knife | ~20%: phones, wallets, bottles, umbrellas, bags | No | 114 images | YOLO + COCO | $149 research / $590 commercial |
| CCTV Fall & Incident Detection | 1,000 images | person, fallen_person | ~20%: sitting, crouching, squatting, kneeling | No | 111 images | YOLO + COCO | $149 research / $590 commercial |
| Airport Drone Threat & Safety | 1,200 images | drone, bird | Birds, as a labelled class | Evaluation clips | 111 images | YOLO | $149 research / $590 commercial |
| CCTV Fight & Aggression Detection | 1,000 images | person, fight | ~28%: hugs, handshakes, dancing, arguments | No | 103 images | YOLO-pose + COCO | $149 research / $590 commercial |
| Pedestrian Attribute Recognition | 10,000 images (50K by quote) | pedestrian (with attributes) | — | No | 996 images | JSONL (Descriptive + Structured) | $149 research / $590 commercial |
| CCTV Weapon Detection: Rifles vs Umbrellas | 1,000 images | person, rifle, umbrella | 50%: umbrellas, as a labelled class | Evaluation clips | 120 images | YOLO | $149 research / $590 commercial |
| Weapon Detection: Handgun vs Bag of Chips | 1,100 images | person, handgun, chip_bag | Chip bags, as a labelled class (balanced split) | Evaluation clips | 110 images | YOLO | $149 research / $590 commercial |
| Military Drone Swarm & Saturation Attack | 1,000+ images | drone | — | No | 117 images | YOLO | $149 research / $590 commercial |
| Drone Civilian / Armed Military Detection | 800 images | civilian, armed_person, vehicle | — | Evaluation clips | On request | YOLO | $149 research / $590 commercial |
| Driver Drowsiness Detection | 10,000 images | eye_open, eye_closed, eye_drowsy | — | Evaluation clips | 168 images | YOLO + Keypoints | $149 research / $590 commercial |
Images is the size of the full set; a standard licence covers 1,000 images, and packs of5,000–20,000 images cost $2,400–$8,500 with a commercial licence. Hard negatives are look-alike objects or scenes that are not the target, included so the model learns what not to flag.
Questions
Before you licence
What is a computer vision dataset?
A labelled collection of images or video used to train and evaluate models that detect, classify or segment objects and events. Each image comes with annotations — bounding boxes, keypoints, masks or attributes — that tell the model what to learn.
What is synthetic computer vision training data?
Imagery generated by rendering and generative models instead of captured by a camera. Because the pipeline knows what it placed in each frame, labels come out with the image. Every Simuletic dataset is 100% synthetic, so it contains no real people.
Which annotation formats do the datasets use?
Detection datasets ship as YOLO (works with YOLOv5, v8 and v11) and COCO. Pose datasets use COCO keypoints, and pedestrian attribute data ships as JSONL. Some datasets also include video, VLM captions or keypoints — see each dataset page.
Can I try a dataset before buying?
Yes. Almost every dataset has a free sample on Kaggle — typically 100 to 400 images — so you can train a baseline and check the labels before licensing the full set.
Can I use the data commercially?
With a commercial licence, yes: one organisation may train models on the data and ship those models in products, perpetually. The research licence covers non-commercial work by one lab or course. You may not resell or redistribute the dataset itself.
Which YOLO versions work with the labels?
The YOLO text format works with YOLOv5, YOLOv8 and YOLO11 out of the box; COCO JSON works with most other detectors, including RT-DETR and Faster R-CNN.
Do any datasets include video?
Yes. Shoplifting, fall detection, airport drones and others ship with video sequences or evaluation clips. Custom datasets can be delivered as video clips with tracking IDs and event times.
What if none of these match my cameras?
Then a custom dataset is usually the better buy: we render the same kinds of events into photos of your own empty scenes, as images or video, with your classes.
What does a licence cost?
Each image dataset is sold as 1,000 images with a one-time licence: $149 for students and non-commercial research, $590 for commercial use. Packs of 5,000–20,000 images are commercial only, $2,400–$8,500. Video datasets are priced on request.
New to synthetic data? Start with what synthetic data for computer vision is.
Not in the catalogue?
We build datasets for scenarios you can't film — on your own camera angles, with your classes and the look-alikes that cause false alarms.