Asymmetric cyclical hashing for large scale image retrieval
Download as DOCX, PDF0 likes146 views
This paper introduces asymmetric cyclical hashing, a novel hashing scheme that uses different hash code lengths for queries and stored images to optimize storage and retrieval efficiency for large scale image databases. By employing a compact hash code for storage and a longer hash code for queries, the method enhances precision and recall in image retrieval. Experimental results demonstrate the effectiveness of this approach on benchmark databases containing up to one million images.
Asymmetric cyclical hashing for large scale image retrieval
- 1. ASYMMETRIC CYCLICAL HASHING FOR LARGE SCALE IMAGE RETRIEVAL ABSTRACT This paper addresses a problem in the hashing technique for large scale image retrieval: learn acompact hash code to reduce the storage cost with performance comparable to that of the long hashcode. A longer hash code yields a better precision rate of retrieved images. However it also requiresa larger storage, which limits the number of stored images. Current hashing methods employ thesame code length for both queries and stored images. We propose a new hashing scheme usingtwo hash codes with different lengths for queries and stored images, i.e., the Asymmetric CyclicalHashing. A compact hash code is used to reduce the storage requirement, while a long hash codeis used for the query image. The image retrieval is performed by computing the Hamming distanceof the long hash code of the query and the cyclically concatenated compact hash code of the storedimage to yield a high precision and recall rate. Experiments on benchmarking databases consistingup to one million images show the effectiveness of the proposed method.