Unsupervised Hashing Models | Awesome Learning to Hash Add your paper to Learning2Hash

Unsupervised Hashing Models

Unsupervised Hashing Models tackle the problem of data retrieval by leveraging the underlying structure of the data without relying on manually provided labels. These models often employ techniques like covariance matrix factorization or clustering to group similar data points. As a result, they achieve a solid balance in retrieval effectiveness, typically performing better than data-independent models but not quite reaching the efficiency of supervised models. However, the computational cost of training, especially due to matrix factorization, can be quite high.

Below is a curated list of key publications on unsupervised hashing models: