AI-07 · Advanced · ENGLISH · PLANNED
Begin with a baseline, trace a model, measure errors and improve it.
Study time is an initial activity estimate, excluding extra reading, revision and extended laboratory or project work. Videos and question bodies have not been produced by adding this outline.
Complete these courses or demonstrate equivalent knowledge. These are recommended prerequisites, not enforced enrolment locks.
YouTube playlist: ARK Artificial Intelligence | Computer Vision
Playlist title reserved in the plan; no YouTube playlist or upload has been created yet.
Sign in to save this learning pathPractical outcome: Inspect how filters respond to a small synthetic image.
Explain pixels and channels using a visual and one worked real-world example. Context: Inspect how filters respond to a small synthetic image. End with a prediction or check-for-understanding.
Explain convolution using a visual and one worked real-world example. Context: Inspect how filters respond to a small synthetic image. End with a prediction or check-for-understanding.
Explain padding using a visual and one worked real-world example. Context: Inspect how filters respond to a small synthetic image. End with a prediction or check-for-understanding.
Explain stride using a visual and one worked real-world example. Context: Inspect how filters respond to a small synthetic image. End with a prediction or check-for-understanding.
Use the first concepts from this section. Inspect how filters respond to a small synthetic image. Produce a labelled diagram, trace or calculation; explain each step and check one changed input.
3–6 questions on the preceding first-half concepts only: recall, visual interpretation, application and one plausible misconception. Explain each answer after submission.
Explain pooling using a visual and one worked real-world example. Context: Inspect how filters respond to a small synthetic image. End with a prediction or check-for-understanding.
Explain receptive fields using a visual and one worked real-world example. Context: Inspect how filters respond to a small synthetic image. End with a prediction or check-for-understanding.
Explain convolutional networks using a visual and one worked real-world example. Context: Inspect how filters respond to a small synthetic image. End with a prediction or check-for-understanding.
Explain transfer learning using a visual and one worked real-world example. Context: Inspect how filters respond to a small synthetic image. End with a prediction or check-for-understanding.
Inspect how filters respond to a small synthetic image. Work without the example, compare two cases, record assumptions and limitations, then self-review correctness, reasoning and clarity.
6–12 questions covering Pixels and channels, Convolution, Padding and the remaining section concepts, with cumulative retrieval from earlier sections. Include a new transfer problem; 70% unlocks progression.
Practical outcome: Compare localisation and classification on an original toy dataset.
Explain image classification using a visual and one worked real-world example. Context: Compare localisation and classification on an original toy dataset. End with a prediction or check-for-understanding.
Explain bounding boxes using a visual and one worked real-world example. Context: Compare localisation and classification on an original toy dataset. End with a prediction or check-for-understanding.
Explain intersection over union using a visual and one worked real-world example. Context: Compare localisation and classification on an original toy dataset. End with a prediction or check-for-understanding.
Explain object detection using a visual and one worked real-world example. Context: Compare localisation and classification on an original toy dataset. End with a prediction or check-for-understanding.
Use the first concepts from this section. Compare localisation and classification on an original toy dataset. Produce a labelled diagram, trace or calculation; explain each step and check one changed input.
3–6 questions on the preceding first-half concepts only: recall, visual interpretation, application and one plausible misconception. Explain each answer after submission.
Explain semantic segmentation using a visual and one worked real-world example. Context: Compare localisation and classification on an original toy dataset. End with a prediction or check-for-understanding.
Explain instance segmentation using a visual and one worked real-world example. Context: Compare localisation and classification on an original toy dataset. End with a prediction or check-for-understanding.
Explain keypoints using a visual and one worked real-world example. Context: Compare localisation and classification on an original toy dataset. End with a prediction or check-for-understanding.
Explain tracking using a visual and one worked real-world example. Context: Compare localisation and classification on an original toy dataset. End with a prediction or check-for-understanding.
Compare localisation and classification on an original toy dataset. Work without the example, compare two cases, record assumptions and limitations, then self-review correctness, reasoning and clarity.
6–12 questions covering Image classification, Bounding boxes, Intersection over union and the remaining section concepts, with cumulative retrieval from earlier sections. Include a new transfer problem; 70% unlocks progression.
Practical outcome: Design a bounded visual-inspection evaluation with clear failure cases.
Explain vision transformers using a visual and one worked real-world example. Context: Design a bounded visual-inspection evaluation with clear failure cases. End with a prediction or check-for-understanding.
Explain self-supervised vision using a visual and one worked real-world example. Context: Design a bounded visual-inspection evaluation with clear failure cases. End with a prediction or check-for-understanding.
Explain image embeddings using a visual and one worked real-world example. Context: Design a bounded visual-inspection evaluation with clear failure cases. End with a prediction or check-for-understanding.
Explain multimodal alignment using a visual and one worked real-world example. Context: Design a bounded visual-inspection evaluation with clear failure cases. End with a prediction or check-for-understanding.
Use the first concepts from this section. Design a bounded visual-inspection evaluation with clear failure cases. Produce a labelled diagram, trace or calculation; explain each step and check one changed input.
3–6 questions on the preceding first-half concepts only: recall, visual interpretation, application and one plausible misconception. Explain each answer after submission.
Explain depth estimation using a visual and one worked real-world example. Context: Design a bounded visual-inspection evaluation with clear failure cases. End with a prediction or check-for-understanding.
Explain robustness to shifts using a visual and one worked real-world example. Context: Design a bounded visual-inspection evaluation with clear failure cases. End with a prediction or check-for-understanding.
Explain vision bias using a visual and one worked real-world example. Context: Design a bounded visual-inspection evaluation with clear failure cases. End with a prediction or check-for-understanding.
Explain visual model explanations using a visual and one worked real-world example. Context: Design a bounded visual-inspection evaluation with clear failure cases. End with a prediction or check-for-understanding.
Design a bounded visual-inspection evaluation with clear failure cases. Work without the example, compare two cases, record assumptions and limitations, then self-review correctness, reasoning and clarity.
6–12 questions covering Vision transformers, Self-supervised vision, Image embeddings and the remaining section concepts, with cumulative retrieval from earlier sections. Include a new transfer problem; 70% unlocks progression.
Combine all three sections: Design a bounded visual-inspection evaluation with clear failure cases. Submit an original worked artifact, evidence or tests, assumptions and a reflection. Rubric: correctness, method, practical usefulness, reproducibility and communication. Self-reviewed, not certified.
6–12 balanced questions sampling every section of Computer Vision, including interpretation, calculation or reasoning, a misconception and a new application. Passing is 70%; review errors before retrying.
Data experiments, evaluations, model tracing and reproducible projects. Guided work leads to an independent attempt, with a diagram, explanation, trace, calculation or tested artifact. Question bodies are generated when opened in your private learning path. Programming submissions are stored as text; the platform does not execute submitted code.
Use original or appropriately licensed datasets. Practical science activities use models or supplied data unless a qualified supervisor has reviewed a physical experiment.
Open the interactive practice sampler →