Training Trains Internship
Student login
← Computer Vision learning path
Computer Vision · Module 7 of 10

Faces, pose and OCR

Apply CV to faces, human pose and text reading.

📖 Key ideas

  1. Face detection finds faces; recognition identifies who.
  2. Pose estimation finds body keypoints (MediaPipe, OpenPose).
  3. OCR reads text: Tesseract, EasyOCR, cloud APIs.
  4. Document AI extracts fields from forms and invoices.
  5. Respect privacy and consent with facial data.
💡 Real-life example: Pose estimation is drawing a stick figure over a person in a photo.
Face recognitionIdentifying a person from a face
KeypointsBody joints detected in pose estimation
TesseractOpen-source OCR engine
MediaPipeGoogle framework for face/hand/pose