True-depth Selfie photo Dataset | 60,000+ IDs| Global Coverage | Face Detection | Computer vision Data |
True‑depth selfie dataset with 60000 IDs (3 files per ID): an RGB face image (jpg), an IR/depth map (png), and an ARKit FaceAnchor file (.face) containing 1220 vertices and 52 blend‑shapes.
Signature Photo Dataset | 3K+ images | 100+ Countries | Identity & Fraud Prevention | Computer Vision Data
Short signature-photo dataset: 3249 clean images of handwritten signatures captured on plain white paper by contributors from 100+ countries. Fully consented. Rich metadata (country, gender, age bands) for training signature verification, writer ID, and document-auth models.
United States
China
Japan
Germany
India
+169 more
Selfie photo Dataset | 10M+ images | Global Coverage | Face Detection | Computer vision Data
Selfie photo dataset with 10M+ JPG images from 10,229,822 unique users (≈1 image per ID) for face detection and broader computer-vision tasks; includes rich metadata.
Selfie Video Dataset | 7340 minutes | 1121 Individuals | 4K & FullHD | Multilingual | Face‑&‑Pose Computer‑Vision Data
Dataset: 1121 selfie videos (≈122h), single standing speaker, horiz, fixed cam. All 1080p+ (39 % 4K, 55 % Full HD), sharp faces, clear light, native monologues, natural expressions, strict QC passed; no occlusions or blur, ideal for face detection, AI-avatar training and CV tasks.
India
Russian Federation
Spain
Kazakhstan
Selfie Video Dataset | 4M+ images | 2M Individuals | Global Coverage | Face Detection | Computer vision Data |
Short selfie‑style video dataset: 4 349 945 clips from 2 060 573 users (3 573 499 videos are ≥3 s, most with voice).
Selfie Video Dataset | 3K+ videos | Global Coverage | Face & Voice Biometrics | Computer-Vision Data
Selfie-style compliance video dataset: 3076 indoor clips (≈20 s each) from about 2403 unique individuals. Ideal for liveness, head-pose, speaker-ID, and multimodal authentication research.
United States
China
Japan
Germany
India
+211 more
Radiology Medical Dataset | 45 M Studies | 10 M Patients | 7 Modalities | Diagnostic AI Data
45 M radiology studies from 10 M de-ID patients (~4.5/ID): X-ray, CT, MRI, mammography, dental, fluoro, PET/SPECT. Each exam carries age/sex, body region, vendor, year, pathology tags. Region-balanced, 2015-25 span. Plug-and-play fuel for disease detection, triage, and multimodal clinical AI.
Belarus
Palm Image Dataset | 78 K+ photos | Global Coverage | Hand Biometrics | Computer‑Vision Data
Data collection: 78 417 images, 11 098 consenting persons (≈7 per subject - rear & selfie cameras ×3 lighting / background setups). Rich metadata tags age, gender, ethnicity, occupation, phone model, dominant hand and jewellery presence - ideal for palm‑based identification, liveness and research.
United States
China
Japan
Germany
India
+169 more
Latin American English Accent Speech Dataset — Authentic Local Speaker Conversations
High-quality English speech dataset with authentic accents from Mexico, Colombia, Dominican Republic, Costa Rica, Guatemala, and El Salvador. Perfect for AI training, accent recognition, and speech research.
Mexico
Colombia
Dominican Republic
Guatemala
Costa Rica
+1 more
ID's photo Dataset | 67 countries | 11 types of documents | Document Recognition | OCR Training | Computer Vision
Dataset of 3623 images from 1661 users (\~2.18/user), mainly front/back ID documents, ideal for OCR training, document recognition, and automated identity verification tasks.
United States
China
India
United Kingdom
France
+62 more