rank.smartfilter

Necessity:optional
Value:off | normal | high | veryhigh | extreme
Default:off

With the param rank.smartfilter you can control the Smart Filter which drastically boosts search performance when searching for similar images. Scoring in Solr is a two step process: collecting relevant documents (number of found documents) and ranking/scoring them afterwards.

Visual similarity search is a compute intensive task and significantly slower than a normal keyword search in Solr. The response time strongly depends on the number of documents that need to be scored.

The Smart Filter is a pre-scoring technique which filters out images that are not likely to be visually similar and relevant. The filter is applied before the expensive visual scoring process happens. This drastically reduces the amount of documents that have to be scored thus improving search performance tremendously while keeping quality loss to a minimum.

Unlike to the Auto Context feature, the Smart Filter does not rely on any metadata. Relevant images are determined via a precalculated cluster descriptor. You can control intensity of filtering and thus performance gain and quality loss. The param rank.smartfilter accepts different filter levels: normal, high, veryhigh and extreme.

An example could look like this:

&pixolution=true&rank.by=url:http://a.com/image.jpg&rank.smartfilter=high