Postgraduate Diploma in Validation of Attention Systems with Computer Vision
800-hour online Postgraduate Diploma in validation of Attention Systems with Computer Vision. 8 modules with Delegated Regulation (EU) 2023/2590, Delegated Regulation (EU) 2021/1341 and OpenCV. Assigned tutor and open enrolment.
Take the controls.
Run a real engineering trade-off from this programme. Move the sliders, watch the model react in real time and beat the challenge.
What you will be able to do
Define measurable requirements for vision-based attention systems.
Build reliable ground truth for gaze and drowsiness.
Design diverse, privacy-compliant validation datasets.
Calculate metrics with confidence intervals and by subgroup.
Assess robustness against edge cases.
Run regulatory and consumer tests.
Maintain validation after model updates.
Who it's for
- Computer vision validation engineer who wants to specialise in Validation of Attention Systems with Computer Vision.
- DMS engineer who wants to specialise in Validation of Attention Systems with Computer Vision.
- Vision data scientist who wants to specialise in Validation of Attention Systems with Computer Vision.
- Annotation and ground truth specialist who wants to specialise in Validation of Attention Systems with Computer Vision.
- Type-approval engineer who wants to specialise in Validation of Attention Systems with Computer Vision.
- Algorithmic fairness analyst who wants to specialise in Validation of Attention Systems with Computer Vision.
- Engineering or science graduates looking to join car manufacturers.
What you'll work with
- Reglamento Delegado (UE) 2023/2590
- Reglamento Delegado (UE) 2021/1341
- ISO 15007
- ISO/IEC 19795-1
- Euro NCAP Safety Assist
- ISO/PAS 8800:2024
- IEC 62471
- Reglamento (UE) 2016/679
8 modules · 800 h
1.1Processing chain: capture, detection and estimation1.2Geometric models and appearance-based gaze models1.3Convolutional networks and transformers in the cabin1.4Operational definition of attention and distraction1.5Gaze zones and regions of interest1.6Required latency and frame rate1.7Sources of error in real environments1.8Functional requirements derived from regulation1.9System validation architecture1.10Case study: measurable requirements for a distraction detector
2.1Eye-tracking reference systems2.2Manual annotation of gaze zones according to ISO 150072.3Inter-annotator agreement and Cohen's kappa2.4Video and reference signal synchronisation2.5Drowsiness annotation with KSS and observers2.6Video annotation tools2.7Annotation quality control2.8Systematic reference errors2.9Annotation costs and scaling2.10Lab: gaze zone annotation with CVAT
3.1Participant sample design3.2Diversity of age, sex, ethnicity and physiognomy3.3Glasses, sunglasses and facial accessories3.4Lighting conditions: day, night and backlight3.5Vehicles, camera positions and seats3.6Public datasets: DMD and Drive&Act3.7Synthetic face data and its validity3.8Biometric data protection3.9Dataset management and versioning3.10Practice: participant sampling plan for a DMS
4.1Gaze accuracy in degrees4.2Confusion matrices by zone4.3False positive and false negative rates4.4ROC curves and threshold selection4.5Event detection latency4.6Event-level versus frame-level metrics4.7ISO/IEC 19795-1 methodology for biometrics4.8Confidence intervals and sample size4.9Subgroup metrics and fairness4.10Workshop: calculating per-event metrics with Python
5.1Infrared lighting variations5.2Partial facial occlusions5.3Extreme head movements5.4Vibration and blur5.5Driver changes during the journey5.6Out-of-distribution detection5.7Stress tests with synthetic perturbations5.8Controlled degradation and unavailable state5.9Robustness reports5.10Practice: robustness test battery with perturbations
6.1ADDW test procedures6.2DDAW test procedures6.3Euro NCAP distraction protocols6.4Standardised distraction scenarios6.5Track tests with real drivers6.6Driving simulator tests6.7Preparing type-approval documentation6.8Repeatability across laboratories6.9Experience from recent assessments6.10Case study: running a Euro NCAP distraction protocol
7.1Validation of new model versions7.2Automated regression tests7.3Field monitoring with privacy7.4Analysis of reported false alarms7.5ISO/PAS 8800 requirements for AI models7.6Change management after OTA updates7.7Fleet quality indicators7.8Periodic fairness audits7.9Dataset improvement7.10Lab: gaze model regression pipeline
8.1Defining measurable requirements8.2Ground truth plan8.3Validation set design8.4Metric selection8.5Robustness testing8.6Simulated regulatory test8.7Fairness analysis8.8Validation report8.9Thesis and presentation8.10Final review with the tutor
How you'll learn
Fully 100% online training on the SEIUM campus, with downloadable material, recorded classes and an assigned technical tutor. No fixed timetables: students progress at their own pace within the intake, making it possible to combine with work.
Each module combines theory, guided reading of the original standard and applied exercises. The working software is OpenCV, PyTorch, CVAT, Tobii Pro Lab, Python with scikit-learn, in free, academic or evaluation versions. Tests are studied using real validation procedures for gaze estimation against reference, robustness tests with image perturbations, and distraction tests in simulator and on track. Assessment is continuous per module plus the final integrative project.
Build it end to end
Your final project follows the real engineering workflow, step by step, reviewed by your tutor.
Define measurable requirements for vision-based attention systems.
Build reliable ground truth for gaze and drowsiness.
Design diverse, privacy-compliant validation datasets.
Calculate metrics with confidence intervals and by subgroup.
Assess robustness against edge cases.
Where this takes you
Requirements
- Prior technical training or professional experience in the sector. No university degree is required.
- Internet connection and a computer capable of running the programme's software.
- Sufficient level of Spanish to follow technical regulatory texts.
Ready to start?
Send your application today. Our admissions team replies within 72 hours with your study plan and funding options.
Frequently asked questions
The programme consists of 800 hours in total, spread over 6 to 8 months. That is between 10 and 15 hours per week on average, compatible with a full-time job.
The course works with OpenCV, PyTorch, CVAT, Tobii Pro Lab, Python with scikit-learn. Free, academic or evaluation versions are used, and the tutor indicates the free alternative where one exists.
An own diploma from SEIUM upon passing the continuous assessment and the integrative project. It is in-house training, not an official Ministry qualification.
Yes. Modules passed in a course or diploma in the same area are credited towards the higher programme. Check the credit recognition policy when enrolling.
Related programmes
Diploma in Regulatory Strategy and International Homologation for AD
Online postgraduate diploma of 800 hours in regulatory strategy and international type approval for automated driving (AD). 8 modules covering the WP.29 framework, UN R157, R155…
Diploma in Urban Operation and Integration with TMS/ATMS
Online postgraduate diploma of 800 hours in urban operation and TMS/ATMS integration. 8 modules covering urban traffic management, autonomous fleets in the city, city–vehicle data exchange,…
Postgraduate Diploma in RL/IL for Driving: Safety Layers and Reward Shaping
Online 800-hour Postgraduate Diploma in RL/IL for Driving: Safety Layers and Reward Shaping. 8 modules covering ISO/PAS 8800:2024, ISO 21448:2022 (SOTIF) and CARLA. Assigned tutor and…
More in Autonomous, ADAS & Software-Defined Vehicle
All Autonomous, ADAS & Software-Defined Vehicle programmes → · How to apply · Fees & financing · Scholarships