Written lessons

Working with RGB-D observations

RGB describes appearance. Depth describes distance according to the sensor's coordinate system and unit convention. Combining them requires attention to calibration, alignment, missing measurements, and scale.

Read more

Training and evaluation

Training adjusts model parameters using examples and a loss function. Validation informs choices such as hyperparameters and stopping time. The test set evaluates the final procedure after those choices are fixed.

Read more

Tensors and shapes

A tensor stores values in a multidimensional array. In image learning, dimensions often describe the batch, channels, height, and width.

Read more