Posts tagged coursera

The YOLO Algorithm, End to End

Every piece of this series assembled into one detector: the grid, the output tensor, the box parameterisation that makes training stable, and the full pipeline.

• cnn
• computer-vision
• coursera
• deep-learning
• Written in EN

Daniel Du
03 Jun 2023 · 5 min read

Anchor Boxes in Object Detection

The one-object-per-cell limit anchor boxes remove, the label vector written out in full, and why k-means on the training set beats hand-picked shapes.

• cnn
• computer-vision
• coursera
• deep-learning
• Written in EN

Daniel Du
01 Jun 2023 · 4 min read

Non-Max Suppression

How detectors collapse dozens of overlapping boxes for one object into a single prediction, the exact algorithm, and the failure it causes in crowds.

• cnn
• computer-vision
• coursera
• deep-learning
• Written in EN

Daniel Du
30 May 2023 · 3 min read

Intersection over Union

The overlap metric every detector is scored on, worked with real coordinates, and why two boxes that look close can still fail the 0.5 threshold.

• cnn
• computer-vision
• coursera
• deep-learning
• Written in EN

Daniel Du
28 May 2023 · 4 min read

The State of Computer Vision

The data-versus-hand-engineering spectrum, why detection sits further along it than classification, and which benchmark tricks not to ship.

• cnn
• computer-vision
• coursera
• deep-learning
• Written in EN

Daniel Du
15 Apr 2023 · 3 min read

Data Augmentation in Computer Vision

The standard augmentations and what invariance each one encodes, PCA colour augmentation written out, and why the wrong transform quietly costs accuracy.

• cnn
• computer-vision
• coursera
• deep-learning
• Written in EN

Daniel Du
14 Apr 2023 · 4 min read