AestheticNet: deep convolutional neural network for person identification from visual aesthetic

Bari, A. S. M. Hossain; Sieu, Brandon; Gavrilova, Marina L.

doi:10.1007/s00371-020-01893-7

Cited by 14 publications

(6 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The most recent work exploited the original deep learning approach [16] for the first time on visual aesthetic-based identification [6]. Above mentioned techniques were highly dependent on manual feature engineering.…”

Section: Literature Reviewmentioning

confidence: 99%

See 1 more Smart Citation

A Novel Three Stage Framework for Person Identification From Audio Aesthetic

Iffath

Gavrilova

2022

IEEE Access

Self Cite

View full text Add to dashboard Cite

Social behavioral biometrics investigates social interactions to determine a person's identity. Within the discipline of social behavioral biometrics, recognition of individuals based on their aesthetic preferences is an emerging direction of research. Human aesthetic is a soft, behavioral biometric trait that refers to a person's attitudes towards a particular subject material. Recent developments in aesthetic-based biometric systems have proven that an individual's visual and audio aesthetic preferences hold considerable distinctive features. This paper introduces a novel three-stage audio-aesthetic system that can uniquely identify a user from the set of their favorite songs. The system utilizes Residual Network (ResNet) for highlevel feature extraction. A hybrid meta-heuristic feature selection algorithm based on Cuckoo Search and Whale Optimization is proposed for feature extraction optimization, which results in the low-dimensional feature set. The selected subset of features is fed into the XGBoost classifier to establish a person's identity. The proposed method outperformed the handcrafted feature-based method by achieving 99.54% accuracy on a proprietary dataset (Free Music Archive) and 99.79% accuracy on a publicly available dataset (Million Playlists Dataset).INDEX TERMS Social behavioral biometrics, deep learning, biometric authentication, audio aesthetics, transfer learning, meta-heuristic, feature selection.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

“…In the domain of social behavioral biometric research, aesthetic systems have shown great potential. Several visual aesthetic systems have been recently developed for person identification [6] and gender prediction [7]. These systems have demonstrated that a user's preferred set of images can hold discriminatory features.…”

Section: Introductionmentioning

confidence: 99%

A Novel Three Stage Framework for Person Identification From Audio Aesthetic

Iffath

Gavrilova

2022

IEEE Access

Self Cite

View full text Add to dashboard Cite

show abstract

“…Visual esthetics is related to the preference of a person that represents their sense of beauty toward images, which can influence perceptions, the trustworthiness of a website, and critically affect user satisfaction and pleasure. Visual esthetic identification can help deeply understand human preference and online social behavior ( Moshagen and Thielsch, 2010 ; Bari et al, 2020 ). In tourism, esthetics and tourism are connected philosophically; in fact, esthetic experience is an essential element of tourism.…”

Section: Theoretical Backgroundmentioning

confidence: 99%

Exploring Destination Choice Intention by Using the Tourism Photographic: From the Perspectives of Visual Esthetic Processing

2021

View full text Add to dashboard Cite

This study aims to deepen the understanding of tourism photography by developing and testing a theoretical model that accounts for the relationships between visual esthetics and destination choice intention of tourists. Thus, we sought to use a stimulus-organism-response (S-O-R) paradigm to predict destination choice intention, which includes three variables related to visual esthetics: first impression, visual appeal, and esthetic emotion. We used the combination of self-reported and eye movement data to examine the cognitive processes of tourists that visual esthetic formation. We found that compared to the built environment and amateur esthetic images, natural environment and professional esthetic images can get (1) higher visual appeal, (2) better first impression, and (3) higher visual processing fluency (or less cognitive effort) and positive esthetic emotions. Furthermore, visual appeal, first impression, and esthetic emotion deriving from environment esthetics and photograph esthetics have a positive impact on destination choice intention. This study has practical implications for destination planning and management.

show abstract

“…Several studies have established the fact that aesthetic preferences can uniquely identify a person 8,9 . Recently developed visual and audio aesthetic‐based biometric systems have revealed that these preferences can be used efficiently for user identification 10,11 . However, the majority of the existing aesthetic biometric systems are unimodal.…”

Section: Introductionmentioning

confidence: 99%

“…8,9 Recently developed visual and audio aesthetic-based biometric systems have revealed that these preferences can be used efficiently for user identification. 10,11 However, the majority of the existing aesthetic biometric systems are unimodal. As a result, such systems are more likely to suffer from limitations such as noise, non-universality, intra-and inter-class variations, and other factors.…”

Section: Introductionmentioning

confidence: 99%

RAIF: A deep learning‐based architecture for multi‐modal aesthetic biometric system

Iffath

Gavrilova

2023

Computer Animation & Virtual

View full text Add to dashboard Cite

Human aesthetics play a significant role in video game development, emotional‐aware robot design, online recommender systems, digital human, and other domains of research focusing on human‐computer interactions. Social network user recognition based on aesthetic preferences is an emerging research domain. In this paper, a novel deep learning architecture is proposed for multi‐modal audio‐visual person identification that combines audio and visual aesthetic features. A pre‐trained ResNet architecture is utilized to extract high‐level features from a set of user‐preferred audio and image samples. A novel deep learning‐based fusion technique called residual‐aided intermediate fusion (RAIF) is introduced in order to effectively merge the audio and visual features. The proposed RAIF method achieved an accuracy of 98% and a loss of 0.01 on a proprietary multi‐modal dataset, indicating its effectiveness in fusing audio and visual information.

show abstract

AestheticNet: deep convolutional neural network for person identification from visual aesthetic

Cited by 14 publications

References 26 publications

A Novel Three Stage Framework for Person Identification From Audio Aesthetic

A Novel Three Stage Framework for Person Identification From Audio Aesthetic

Exploring Destination Choice Intention by Using the Tourism Photographic: From the Perspectives of Visual Esthetic Processing

RAIF: A deep learning‐based architecture for multi‐modal aesthetic biometric system

Contact Info

Product

Resources

About