Learn
The blog lists everything by date. These two arrange the same articles by what you need first, so you can read straight down instead of guessing where to start.
Computer Vision Fundamentals
Imaging, image processing, features, geometry. 29 articles, ordered by dependency: nothing appears before the thing it depends on.
Imaging
How light becomes an array of numbers.
How a scene is flattened onto a sensor, and what the glass does on the way.
What the number in a pixel counts, and how far it can be trusted.
What the array holds, where its samples sit, how finely they are written, and what the file did to them.
Image processing
What you do to that array once you have it.
The one operation behind blur, sharpening, edges and every CNN.
Features
What in an image is worth pointing at, and how to describe it.
Finding where one thing stops and another begins.
Surviving a camera that moves: scale space, and 128 numbers per point.
Geometry
From one camera to many, and back out to three dimensions.
Intrinsics, extrinsics, calibration, and pose from known points.
The Computer Vision Metrics Handbook
Every score in computer vision, from the ground up. A self-contained book on evaluation with its own reading order: one confusion matrix grows into precision and recall, and those stretch over boxes, masks, identities through time, and images with no ground truth at all.
The handbook has its own page ›
- 01ClassificationClassification Metrics, From the Ground Up: Precision, Recall, F1, ROC and AUC
The confusion matrix and everything that falls out of it: precision, recall, F1, specificity, the precision/recall trade-off, PR and ROC curves, and AUC, all built on one running toy example (a backyard cat-cam), with the equations, the intuition, and scikit-learn.
Confusion matrixPrecisionRecallF1SpecificityPR & ROC curvesAUC - 02Multiclass & multilabelMulticlass and Multilabel Metrics: Micro, Macro, Weighted, and the Difference That Matters
How precision/recall/F1 scale past two classes: one-vs-rest, and the micro/macro/weighted averaging that decides whether rare classes count. Plus multilabel-specific metrics: subset accuracy, Hamming loss, and top-k, all on one running shelter-camera example, with scikit-learn.
One-vs-restMicroMacroWeightedSubset accuracyHamming lossTop-k - 03Object detectionObject Detection Metrics: IoU, AP, and the mAP Everyone Quotes
How detection is scored: IoU to decide a hit, greedy matching into TP/FP/FN, per-class Average Precision as the area under the precision–recall curve, and the COCO mAP@[.5:.95] you see in every paper, all on one running parking-lot example, with torchmetrics.
IoUAPPR curvemAP - 04SegmentationSegmentation Metrics: IoU, Dice, mIoU, and Panoptic Quality
Scoring masks instead of boxes: pixel IoU/Jaccard and the Dice coefficient (and why they're related), mIoU for semantic segmentation, mask-AP for instance segmentation, and Panoptic Quality for the unified task, on one running cat-mask example, with the equations and code.
IoU / JaccardDicemIoUMask APPanoptic Quality - 05Object trackingObject Tracking Metrics: MOTA, MOTP, IDF1, and HOTA
Tracking adds identity over time, so its metrics score two things at once: detecting objects and keeping their IDs consistent. MOTA, MOTP, ID switches, IDF1, and the modern HOTA that balances detection against association, on one running door-cam example, with the equations and motmetrics.
MOTAMOTPID switchesIDF1HOTA - 06Image generationImage Generation Metrics: FID, IS, KID, LPIPS, and CLIPScore
How do you score images with no ground truth? You compare distributions and use learned perceptual models: Inception Score, Fréchet Inception Distance, KID, LPIPS, and CLIPScore for prompt fidelity, on one running cat-generator example, with the equations, the catches, and torchmetrics.
Inception ScoreFIDKIDLPIPSCLIPScore
Model and edge-deployment articles have no reading order in the same way, since each one is a benchmark that stands alone, so they live on the blog.