Computer Vision, Imagenet : Terms confusion

  1. Is it correct to say that the sensing and interpreting devices in the picture represent computer vision?


  1. If correct, are the layers in the picture containing blue and purple neurons representing the tasks carried out by computer vision?

  2. Is it correct to say that the tool or software used for this computer vision uses the library of millions of images created by imagenet for object identification?

waiting for response

Computer vision refers to the ability of computers and systems to extract, analyze, and understand information from visual data – essentially, enabling computers to ‘see’ and interpret the world similarly to how humans do. This involves capturing, processing, analyzing, and making sense of visual data from the surrounding environment. In essence, computer vision refers to the ability of computers to “see” things the way humans do (though the accuracy may not be comparable to humans).

Not really. In Computer vision, Each layer in a neural network processes the input data in a specific way. Early layers might detect simple features like edges or colors, while deeper layers can identify more complex patterns or objects. These layers work together to interpret visual data – such as images or videos – and extract meaningful information.

Yeah. Just that imagenet is not the only resource available.

First, what exactly do you mean by “computer vision”? It’s a very broad term

The sensing device is just a camera. It does nothing specific to any algorithm.

Maybe. You’ve drawn a general-purpose fully-connected neural network. It can be used for lots of tasks - including image recognition.

Not necessarily. There isn’t any “library” necessarily. A neural network may be trained on a lot of images, but the images are not a part of the completed model.

“imagenet” is just one set of data.

1 Like

computer vision application for face recognition


Face recognition is rather more complicated than a single fully-connected NN can accomplish.

It is covered in some detail in the Deep Learning Specialization.

In course 1 of deep learning specialization or later?

Course 4 of Deep Learning Specialization.

Called “Convolutional Neural Networks”.

1 Like