Designed and developed an image search engine using Hadoop Inverted Indexes and Lucene indexing which returns the images pertaining to the input query by the user. Employed FiftyOne API - an object detection software - to detect objects in an image and use the indexed objects called annotations to match with the query. These images are ranked for relevancy using the area of bounding boxes of each annotation and then returned to the frontend to be displayed on the browser.
Search query: Bear mode: Hadoop with Area Ranking
Search query: Motorcycle mode: Hadoop with Area Ranking
Search query: Motorcycle mode: Lucene with TF-IDF Ranking
