Skip to main content

Deep Learning based CBIR 101 (Part 1): intro

Content Based Image Retrieval (CBIR) systems are aimed to find similar images to an input image by querying an image data set. I am pretty sure you've used a CBIR several times before: here's the most popular one:



The search in CBIRs isn't text-based as images are indexed by their visual content. Features are extracted from an image database and stored in some file system or object storage.Then, any time a new image is passed as search criterion, all those images which features are closest to the input image features are returned as results.
The traditional feature extraction process relies on algorithms such as SURF or SIFT, but in this series I am going to describe my journey into CBIR using an unsupervised Deep Learning approach.
Here's a preview of the first attempt, in order to search for food images (the input image in this case is an apple pie):



In the next posts of this series I will walk through the details of the implementation before making things more complex by considering more challenging image sets. Stay tuned!

Labels:

Comments

Post a Comment

Popular posts from this blog

Shipping and analysing MongoDB logs using the Streamsets Data Collector, ElasticSearch and Kibana

In order to show that the considerations done in my last post are general for any log shipping purpose, let's see now how the same process applies to a more real use case scenario: the log shipping and analysis of a MongoDB database logs. MongoDB logs pattern Starting from the release 3.0 (I am considering the release 3.2 for this post) the MongoDB logs come with the following pattern: <timestamp> <severity> <component> [<context>] <message> where:     timestamp is in iso8601-local format.     severity is the level associated to each log message. It is a single character field. Possible values are F (Fatal), E (Error), W (Warning), I (Informational) and D (Debug).     component is for a functional categorization of the log message. Please refer to the specific release of MongoDB you're using to know the full list of possible values.     context is the specific context for a me...
Post a Comment
Read more

Load testing MongoDB using JMeter

Apache JMeter ( http://jmeter.apache.org/ ) added support for MongoDB since its 2.10 release. In this post I am referring to the latest JMeter release (2.13). A preliminary JMeter setup is needed before starting your first test plan for MongoDB. It uses Groovy as scripting reference language, so Groovy needs to be set up for our favorite load testing tool. Follow these steps to complete the set up: Download Groovy from the official website ( http://www.groovy-lang.org/download.html ). In this post I am referring to the Groovy release 2.4.4, but using later versions is fine. Copy the groovy-all-2.4.4.jar to the $JMETER_HOME/lib folder. Restart JMeter if it was running while adding the Groovy JAR file. Now you can start creating a test plan for MongoDB load testing. From the UI select the MongoDB template ( File -> Templates... ). The new test plan has a MongoDB Source Config element. Here you have to setup the connection details for the database to be tested: The Threa...
43 comments
Read more

Turning Python Scripts into Working Web Apps Quickly with Streamlit

 I just realized that I am using Streamlit since almost one year now, posted about in Twitter or LinkedIn several times, but never wrote a blog post about it before. Communication in Data Science and Machine Learning is the key. Being able to showcase work in progress and share results with the business makes the difference. Verbal and non-verbal communication skills are important. Having some tool that could support you in this kind of conversation with a mixed audience that couldn't have a technical background or would like to hear in terms of results and business value would be of great help. I found that Streamlit fits well this scenario. Streamlit is an Open Source (Apache License 2.0) Python framework that turns data or ML scripts into shareable web apps in minutes (no kidding). Python only: no front‑end experience required. To start with Streamlit, just install it through pip (it is available in Anaconda too): pip install streamlit and you are ready to execute the working de...
18 comments
Read more