Cross-modal hashing is the compression of high dimensional data into compact binary codes with similar binary codes for similar objects. The key ideas is to create codes for cross-modal retrieval. The constraints are :

- It has to be an N-dimensional hamming space – a binary representation with a controllable number of bits.
- The same object from different modalities has to have a similar hash code
- The space has to be similarity-preserving

(Baltru\vsaitis et al., 2017) has some good reference papers.

# Bibliography

Baltru\vsaitis, Tadas, Ahuja, C., & Morency, L., *Multimodal machine learning: a survey and taxonomy*, CoRR, *()*, (2017). ↩