IICTI Certified Computer Vision Engineer

Holders demonstrate that they can apply and evaluate vision backbones, foundation models and architecture selection, detection, segmentation and multi-object tracking systems, multi-view geometry, 3D reconstruction and neural rendering and vision-language and open-vocabulary multimodal models, plus 7 further areas, in realistic professional scenarios, as defined in the AQSCERT scheme for Computer Vision Engineer.

Who it is for: Professionals working, or preparing to work, in artificial intelligence roles.

1. 01Vision backbones, foundation models and architecture selection8% of the exam

   • Evaluate and select vision architectures and foundation models that fit task, data and deployment constraints.
   • Scope: CNNs versus Vision Transformers, ConvNeXt, DINOv2, CLIP and SigLIP, SAM 2, accuracy, latency and data-efficiency trade-offs.
2. 02Detection, segmentation and multi-object tracking systems15% of the exam

   • Design end-to-end perception pipelines for detection, segmentation and tracking with appropriate metrics.
   • Scope: DETR and RT-DETR, YOLO family, Mask2Former, panoptic segmentation, ByteTrack and re-identification, mAP, PQ and HOTA metrics.
3. 03Multi-view geometry, 3D reconstruction and neural rendering11% of the exam

   • Architect 3D perception and reconstruction solutions grounded in sound geometric modelling.
   • Scope: Camera calibration, epipolar geometry, structure-from-motion (COLMAP), visual SLAM, monocular depth, NeRF and 3D Gaussian Splatting.
4. 04Vision-language and open-vocabulary multimodal models6% of the exam

   • Design multimodal vision systems that leverage language supervision while controlling reliability risks.
   • Scope: Vision-language models, Grounding DINO and OWL-ViT, visual grounding and captioning, multimodal evaluation and hallucination analysis.
5. 05Data-centric pipelines, annotation quality and synthetic data10% of the exam

   • Lead data strategies that systematically improve model performance through dataset quality and coverage.
   • Scope: Labelling QA and inter-annotator agreement, active learning, dataset versioning, domain randomisation and diffusion-based synthetic data, datasheets.
6. 06Large-scale training, self-supervision and efficient fine-tuning12% of the exam

   • Design training regimes that achieve target accuracy efficiently at scale.
   • Scope: Self-supervised pretraining, transfer learning, LoRA and PEFT, DDP and FSDP, mixed precision, augmentation policies and hyperparameter optimisation.
7. 07Evaluation, robustness and failure analysis10% of the exam

   • Evaluate vision models rigorously and diagnose failure modes before and after deployment.
   • Scope: Test set design, slice-based error analysis, calibration, out-of-distribution detection, distribution shift and adversarial robustness evaluation.
8. 08Edge and real-time inference optimisation9% of the exam

   • Optimise vision models and pipelines to meet real-time latency, throughput and power targets on edge hardware.
   • Scope: INT8 and FP8 quantisation, pruning and distillation, TensorRT, ONNX Runtime and OpenVINO, DeepStream and GStreamer video pipelines, latency budgeting.
9. 09Video understanding and spatiotemporal analytics6% of the exam

   • Design video analytics systems that reason reliably over time and across camera networks.
   • Scope: Action recognition, video transformers, temporal action localisation, streaming inference and multi-camera analytics.
10. 10Vision MLOps and production monitoring7% of the exam

    • Lead the operation of production vision systems with continuous monitoring and controlled model updates.
    • Scope: Model registries, CI/CD for models, shadow and canary deployment, data and concept drift monitoring, human-in-the-loop feedback and retraining triggers.
11. 11Responsible, secure and compliant computer vision6% of the exam

    • Evaluate vision systems for legal, ethical and security risk and design proportionate safeguards.
    • Scope: EU AI Act biometric and high-risk obligations, GDPR special-category biometric data, fairness auditing, privacy-preserving redaction, adversarial patch threats, ISO/IEC 42001.

Question types

Single choice 100%

Delivery

Online (remote proctored), Test centre, On-site (corporate)

Retakes

Up to three attempts within any 12-month period. A 14-day waiting period applies after a first unsuccessful attempt and 30 days after a second. Each retake requires a new booking and fee. Retakes draw a different randomised form from the question bank.

Provisional cut score — psychometric validation is required before live delivery. Blueprint v1.0.

1. 1

   Register

   Create a free account and start the programme.

2. 2

   Practice

   Online practice test (coming soon).

3. 3

   Training

   Optional

   Online course or classroom — or skip it.

4. 4

   Certification

   Independent exam and certification decision.

Training is optional and does not by itself confer certification. Find a training partner