Hi Kevin,

I've been using Lire for a little while now and was happy to see a plugin 
for ES had been created. Great work. 

I have a question about the score in ES vs the score in Lire.

For the same dataset, indexed with the same features, and then searched 
with the same feature I get the same results however, there are different 
scores in the ES results from those in the Lire results, is this normal?

I'm asking because I have some historical stuff which has recorded the 
scores given by Lire, and the lower ES scores are going to make it 
problematic to use the plugin.

An example:

ES top 3 scores:
0.06624806
0.04820513
0.036607143

Lire Jar top 3 scores:
0.8094672560691833
0.7381518483161926
0.6551922559738159


I'm keen to use ES as it helps to solve some scaling problems. But I'm very 
new to ES so I'm not sure how to look in to the difference with the 
scoring, do you have any insight?

Many thanks,
V



On Thursday, March 20, 2014 10:41:20 AM UTC, Kevin Wang wrote:
>
> Hi All,
>
> I've released version 1.2.0 of Elasticsearch Image Plugin.
> The Image Plugin is an Content Based Image Retrieval Plugin for 
> Elasticsearch using LIRE (Lucene Image Retrieval). It allows users to index 
> images and search for similar images.
>
> Changes in 1.2.0:
>
>    - Use multi-thread when multiple features are required to improve 
>    index speed
>    - Allow index metadata
>    - Allow query by existing image in index
>    
>
>
> https://github.com/kzwang/elasticsearch-image
>
> Also I've created a demo website for this plugin (
> http://demo.elasticsearch-image.com/), it has 1,000,000 images  from 
> MIRFLICKR-1M collection (http://press.liacs.nl/mirflickr)
>
>
> Thanks,
> Kevin
>
>

-- 
You received this message because you are subscribed to the Google Groups 
"elasticsearch" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/elasticsearch/a387f60e-63f0-44d8-9960-6caa918c97ca%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to