Discipline 04 of 09
Systems thatsee and learn
We build practical artificial intelligencefrom computer vision that inspects and counts and recognizesto models that predict and classify and automate decisions

What we engineer
Artificial Intelligencein practice
What RoboFab engineers in Artificial Intelligence
- 01
Computer vision
Detection and tracking and measurement and recognition from cameras in real environments
- 02
Object detection
Custom trained detectors for products and defects and people and vehicles
- 03
Machine learning
Prediction and classification models built on your data and validated honestly
- 04
Deep learning
Neural networks for images and signals and sequences when classic methods fall short
- 05
Edge AI deployment
Models optimized to run on embedded hardware right next to the sensor
- 06
Intelligent automation
AI connected to machines and dashboards and workflows so insight becomes action
Typical architecture
How the systemfits together
A common shape for this kind of systemEvery project gets its own architecture
Typical Artificial Intelligence architecture
Vision and learning systems that turn data into decisions
Typical applications
- Quality inspection
- Face and object recognition
- People and vehicle counting
- Gesture control
- Predictive maintenance
- Smart assistants
Technologies
- Python
- OpenCV
- YOLO
- PyTorch
- TensorFlow
- MediaPipe
- Dataset labeling
- Model optimization
- Raspberry Pi
You receive
- Trained models
- Dataset and labeling guide
- Inference application
- Accuracy report
Reference visual
Reference visual
Reference visual
Reference visual
Reference visuals illustrate the field and are not RoboFab project records
Engineering studies
How we wouldapproach it
Concept studies that show our methodThey are illustrations and not delivered client projects
Related engineering studies
Engineering studyReference visual
Robotics and AutomationArtificial Intelligence
Vision Guided Sorting Cell
A camera and a trained model decide where every part belongs
Read the study Engineering studyReference visual
Artificial IntelligenceEmbedded Systems
Edge AI Inspection Camera
Defect detection that runs on the device without a PC on the line
Read the study Engineering studyReference visual
Embedded SystemsSoftware Engineering
Smart Access and Attendance
Doors and attendance managed from one secure dashboard
Read the study
Works closely with
Connecteddisciplines
Related disciplines
Reference visual01
Embedded Systems
Firmware and controllers that make hardware think
ESP32STM32ArduinoRaspberry Pi
Explore the discipline Reference visual05
Software Engineering
Web platforms and dashboards and APIs that run the business
TypeScriptReactPythonPostgreSQL
Explore the discipline Reference visual03
Robotics and Automation
Machines that move and sense and decide
Servo controlStepper controlPID controlInverse kinematics
Explore the discipline
