From Rank-N to Rank-1 Face Recognition Based on Motion Similarity

Hui Fang, Nick Costen

Research output: Contribution to conferencePoster

6 Citations (Scopus)
3 Downloads (Pure)

Abstract

In this paper, we present a sequential framework using facial motion information as a subsidiary to improve face recognition performance. As is generally known, reasonable static face recognition has been achieved based on subspace reduction techniques. In order to further improve performance, some extra cues, such as temporal variation, are investigated by building dynamic models. We propose a permuted similarity motion feature and integrate it into a sequential recognition system. This system can select the best candidate from the Rank-N candidates picked up in the recognition step based on static appearance parameters by using motion information. The recognition rate of the motion similarity is compared with the motion feature obtained from auto-regressive models to prove its efficiency. In addition, the sequential system achieves better performance when the motion information is integrated with the static appearance information in a flexible manner.
Original languageEnglish
Pages1-10
Publication statusPublished - 2009
EventBritish Machine Vision Conference - London, United Kingdom
Duration: 1 Jan 2009 → …

Conference

ConferenceBritish Machine Vision Conference
CountryUnited Kingdom
CityLondon
Period1/01/09 → …

Fingerprint

Face recognition
Dynamic models

Cite this

Fang, H., & Costen, N. (2009). From Rank-N to Rank-1 Face Recognition Based on Motion Similarity. 1-10. Poster session presented at British Machine Vision Conference, London, United Kingdom.
Fang, Hui ; Costen, Nick. / From Rank-N to Rank-1 Face Recognition Based on Motion Similarity. Poster session presented at British Machine Vision Conference, London, United Kingdom.
@conference{830147d5e7d441f99b87004580229dab,
title = "From Rank-N to Rank-1 Face Recognition Based on Motion Similarity",
abstract = "In this paper, we present a sequential framework using facial motion information as a subsidiary to improve face recognition performance. As is generally known, reasonable static face recognition has been achieved based on subspace reduction techniques. In order to further improve performance, some extra cues, such as temporal variation, are investigated by building dynamic models. We propose a permuted similarity motion feature and integrate it into a sequential recognition system. This system can select the best candidate from the Rank-N candidates picked up in the recognition step based on static appearance parameters by using motion information. The recognition rate of the motion similarity is compared with the motion feature obtained from auto-regressive models to prove its efficiency. In addition, the sequential system achieves better performance when the motion information is integrated with the static appearance information in a flexible manner.",
author = "Hui Fang and Nick Costen",
year = "2009",
language = "English",
pages = "1--10",
note = "British Machine Vision Conference ; Conference date: 01-01-2009",

}

Fang, H & Costen, N 2009, 'From Rank-N to Rank-1 Face Recognition Based on Motion Similarity' British Machine Vision Conference, London, United Kingdom, 1/01/09, pp. 1-10.

From Rank-N to Rank-1 Face Recognition Based on Motion Similarity. / Fang, Hui; Costen, Nick.

2009. 1-10 Poster session presented at British Machine Vision Conference, London, United Kingdom.

Research output: Contribution to conferencePoster

TY - CONF

T1 - From Rank-N to Rank-1 Face Recognition Based on Motion Similarity

AU - Fang, Hui

AU - Costen, Nick

PY - 2009

Y1 - 2009

N2 - In this paper, we present a sequential framework using facial motion information as a subsidiary to improve face recognition performance. As is generally known, reasonable static face recognition has been achieved based on subspace reduction techniques. In order to further improve performance, some extra cues, such as temporal variation, are investigated by building dynamic models. We propose a permuted similarity motion feature and integrate it into a sequential recognition system. This system can select the best candidate from the Rank-N candidates picked up in the recognition step based on static appearance parameters by using motion information. The recognition rate of the motion similarity is compared with the motion feature obtained from auto-regressive models to prove its efficiency. In addition, the sequential system achieves better performance when the motion information is integrated with the static appearance information in a flexible manner.

AB - In this paper, we present a sequential framework using facial motion information as a subsidiary to improve face recognition performance. As is generally known, reasonable static face recognition has been achieved based on subspace reduction techniques. In order to further improve performance, some extra cues, such as temporal variation, are investigated by building dynamic models. We propose a permuted similarity motion feature and integrate it into a sequential recognition system. This system can select the best candidate from the Rank-N candidates picked up in the recognition step based on static appearance parameters by using motion information. The recognition rate of the motion similarity is compared with the motion feature obtained from auto-regressive models to prove its efficiency. In addition, the sequential system achieves better performance when the motion information is integrated with the static appearance information in a flexible manner.

M3 - Poster

SP - 1

EP - 10

ER -

Fang H, Costen N. From Rank-N to Rank-1 Face Recognition Based on Motion Similarity. 2009. Poster session presented at British Machine Vision Conference, London, United Kingdom.