scabini / RADAMView on GitHub
We propose a new method named Random encoding of Aggregated Deep Activation Maps (RADAM) for feature extraction from pre-trained Deep CNNs. The technique consists of encoding the output at different depths of the CNN using a Randomized Autoencoder, producing a single image descriptor
39Dec 17, 2024Updated last year

Alternatives and similar repositories for RADAM

Users that are interested in RADAM are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.

Sorting:

Are these results useful?