Can I read Compact Image Descriptor Aggregation on EtoBox?

Compact Image Descriptor Aggregation by 弘元 is a document available to read on EtoBox.

What is Compact Image Descriptor Aggregation about?

This document proposes a new method called VLAD (Vector of Locally Aggregated Descriptors) to compactly represent images using local descriptors for large-scale image search. VLAD aggregates local descriptors into a vector representation that achieves search accuracy comparable to bag-of-features (BOF) representations but using only 20 bytes per image. The method jointly optimizes the dimensionality reduction and indexing algorithm to best preserve vector similarity comparisons while improving efficiency an

Author

弘元

Language

EN