Sorry, you need to enable JavaScript to visit this website.

facebooktwittermailshare

Vector compression for similarity search using Multi-layer Sparse Ternary Codes

Abstract: 

It was shown recently that Sparse Ternary Codes (STC) posses superior ``coding gain'' compared to the classical binary hashing framework and can successfully be used for large-scale search applications. This work extends the STC for compression and proposes a rate-distortion efficient design. We first study a single-layer setup where we show that binary encoding intrinsically suffers from poor compression quality while STC, thanks to the flexibility in design, can have near-optimal rate allocation. We further show that single-layer codes should be limited to very low rates. Therefore, in order to target arbitrarily high rates, we adopt a multi-layer solution inspired by the classical idea of residual quantization. The proposed architecture, while STC in nature and hence suitable for similarity search, can add the ``list-refinement'' technique as a useful element to the similarity search setup. This can be achieved thanks to the excellent rate-distortion performance of this scheme which we validate on synthetic, as well as large-scale public databases.

up
0 users have voted:

Paper Details

Authors:
Sohrab Ferdowsi, Slava Voloshynovskiy, Dimche Kostadinov
Submitted On:
1 June 2018 - 12:45pm
Short Link:
Type:
Poster
Event:
Presenter's Name:
Sohrab Ferdowsi
Paper Code:
1080
Document Year:
2018
Cite

Document Files

DSW2018_poster.pdf

(18 downloads)

Keywords

Additional Categories

Subscribe

[1] Sohrab Ferdowsi, Slava Voloshynovskiy, Dimche Kostadinov, "Vector compression for similarity search using Multi-layer Sparse Ternary Codes", IEEE SigPort, 2018. [Online]. Available: http://sigport.org/3229. Accessed: Jun. 20, 2018.
@article{3229-18,
url = {http://sigport.org/3229},
author = {Sohrab Ferdowsi; Slava Voloshynovskiy; Dimche Kostadinov },
publisher = {IEEE SigPort},
title = {Vector compression for similarity search using Multi-layer Sparse Ternary Codes},
year = {2018} }
TY - EJOUR
T1 - Vector compression for similarity search using Multi-layer Sparse Ternary Codes
AU - Sohrab Ferdowsi; Slava Voloshynovskiy; Dimche Kostadinov
PY - 2018
PB - IEEE SigPort
UR - http://sigport.org/3229
ER -
Sohrab Ferdowsi, Slava Voloshynovskiy, Dimche Kostadinov. (2018). Vector compression for similarity search using Multi-layer Sparse Ternary Codes. IEEE SigPort. http://sigport.org/3229
Sohrab Ferdowsi, Slava Voloshynovskiy, Dimche Kostadinov, 2018. Vector compression for similarity search using Multi-layer Sparse Ternary Codes. Available at: http://sigport.org/3229.
Sohrab Ferdowsi, Slava Voloshynovskiy, Dimche Kostadinov. (2018). "Vector compression for similarity search using Multi-layer Sparse Ternary Codes." Web.
1. Sohrab Ferdowsi, Slava Voloshynovskiy, Dimche Kostadinov. Vector compression for similarity search using Multi-layer Sparse Ternary Codes [Internet]. IEEE SigPort; 2018. Available from : http://sigport.org/3229