Using NoSQL for a Visual Recognition Service

Catchoom is a technology startup founded in 2011 in Barcelona. Our platform provides visual recognition as a service, it’s designed for applications that need to search through a database of images. Catchoom is empowering leading Augmented Reality companies like Layar and Metaio with scalable and fast visual search.

This presentation will explain the architecture of the Catchoom Platform, emphasizing on the advantages of using NoSQL for the specific requirements of visual recognition.

1) Introduction
– David Arcos
– Catchoom
– Customers
– Visual Recognition
– Examples
– Catchoom Recognition Service
– Team

2) What did we need?
– Minimum requirements
– Flexible
– Reliable
– Fast

3) How we build it
– Technology stack
– The Panel
– Mobile apps
– Data models
– Components
– NoSQL features

4) Advantages of NoSQL
– Performance
– Scalability
– Unstructured data: query
– Unstructured data: metadata
– Availability

5) Cool uses of NoSQL
– Real time calculations (stats, monitoring, rate limiting)
– Sorted sets (indexing)
– Cache
– Volatile data
– Messages (pub/sub, push/pop)

6) Limits
– Django apps compatibility

7) Conclusion
– Summary
– Recommendations
– Thanks for attending
– Q&A

Video producer:
Slides of the presentation:

Related Content:

  • No Related Content