Hands in Focus: Sign Language Recognition Via Top-Down Attention

Publikation: Beiträge in SammelwerkenAufsätze in KonferenzbändenForschungbegutachtet

Standard

Hands in Focus: Sign Language Recognition Via Top-Down Attention. / Sarhan, Noha; Wilms, Christian; Closius, Vanessa et al.
2023 IEEE International Conference on Image Processing, ICIP 2023 - Proceedings: Proceedings. Piscataway: IEEE Electromagnetic Compatibility Society, 2023. S. 2555-2559 (Proceedings - International Conference on Image Processing, ICIP).

Publikation: Beiträge in SammelwerkenAufsätze in KonferenzbändenForschungbegutachtet

Harvard

Sarhan, N, Wilms, C, Closius, V, Brefeld, U & Frintrop, S 2023, Hands in Focus: Sign Language Recognition Via Top-Down Attention. in 2023 IEEE International Conference on Image Processing, ICIP 2023 - Proceedings: Proceedings. Proceedings - International Conference on Image Processing, ICIP, IEEE Electromagnetic Compatibility Society, Piscataway, S. 2555-2559, 2023 IEEE International Conference on Image Processing, Kuala Lumpur, Malaysia, 08.10.23. https://doi.org/10.1109/icip49359.2023.10222729

APA

Sarhan, N., Wilms, C., Closius, V., Brefeld, U., & Frintrop, S. (2023). Hands in Focus: Sign Language Recognition Via Top-Down Attention. In 2023 IEEE International Conference on Image Processing, ICIP 2023 - Proceedings: Proceedings (S. 2555-2559). (Proceedings - International Conference on Image Processing, ICIP). IEEE Electromagnetic Compatibility Society. https://doi.org/10.1109/icip49359.2023.10222729

Vancouver

Sarhan N, Wilms C, Closius V, Brefeld U, Frintrop S. Hands in Focus: Sign Language Recognition Via Top-Down Attention. in 2023 IEEE International Conference on Image Processing, ICIP 2023 - Proceedings: Proceedings. Piscataway: IEEE Electromagnetic Compatibility Society. 2023. S. 2555-2559. (Proceedings - International Conference on Image Processing, ICIP). doi: 10.1109/icip49359.2023.10222729

Bibtex

@inbook{8aefa664c0814a45889ac286322e809a,
title = "Hands in Focus: Sign Language Recognition Via Top-Down Attention",
abstract = "In this paper, we propose a novel Sign Language Recognition (SLR) model that leverages the task-specific knowledge to incorporate Top-Down (TD) attention to focus the processing of the network on the most relevant parts of the input video sequence. For SLR, this includes information about the hands' shape, orientation and positions, and motion trajectory. Our model consists of three streams that process RGB, optical flow and TD attention data. For the TD attention, we generate pixel-precise attention maps focusing on both hands, thereby retaining valuable hand information, while eliminating distracting background information. Our proposed method outperforms state-of-the-art on a challenging large-scale dataset by over 2%, and achieves strong results with a much simpler architecture compared to other systems on the newly released AUTSL dataset [1].",
keywords = "Informatics, sign language recognition, top-down attention, deep learning",
author = "Noha Sarhan and Christian Wilms and Vanessa Closius and Ulf Brefeld and Simone Frintrop",
note = "Publisher Copyright: {\textcopyright} 2023 IEEE.; 2023 IEEE International Conference on Image Processing, ICIP 2023 ; Conference date: 08-10-2023 Through 11-10-2023",
year = "2023",
month = oct,
day = "8",
doi = "10.1109/icip49359.2023.10222729",
language = "English",
isbn = "978-1-7281-9836-1",
series = "Proceedings - International Conference on Image Processing, ICIP",
publisher = "IEEE Electromagnetic Compatibility Society",
pages = "2555--2559",
booktitle = "2023 IEEE International Conference on Image Processing, ICIP 2023 - Proceedings",
url = "https://2023.ieeeicip.org/",

}

RIS

TY - CHAP

T1 - Hands in Focus: Sign Language Recognition Via Top-Down Attention

AU - Sarhan, Noha

AU - Wilms, Christian

AU - Closius, Vanessa

AU - Brefeld, Ulf

AU - Frintrop, Simone

N1 - Conference code: 30

PY - 2023/10/8

Y1 - 2023/10/8

N2 - In this paper, we propose a novel Sign Language Recognition (SLR) model that leverages the task-specific knowledge to incorporate Top-Down (TD) attention to focus the processing of the network on the most relevant parts of the input video sequence. For SLR, this includes information about the hands' shape, orientation and positions, and motion trajectory. Our model consists of three streams that process RGB, optical flow and TD attention data. For the TD attention, we generate pixel-precise attention maps focusing on both hands, thereby retaining valuable hand information, while eliminating distracting background information. Our proposed method outperforms state-of-the-art on a challenging large-scale dataset by over 2%, and achieves strong results with a much simpler architecture compared to other systems on the newly released AUTSL dataset [1].

AB - In this paper, we propose a novel Sign Language Recognition (SLR) model that leverages the task-specific knowledge to incorporate Top-Down (TD) attention to focus the processing of the network on the most relevant parts of the input video sequence. For SLR, this includes information about the hands' shape, orientation and positions, and motion trajectory. Our model consists of three streams that process RGB, optical flow and TD attention data. For the TD attention, we generate pixel-precise attention maps focusing on both hands, thereby retaining valuable hand information, while eliminating distracting background information. Our proposed method outperforms state-of-the-art on a challenging large-scale dataset by over 2%, and achieves strong results with a much simpler architecture compared to other systems on the newly released AUTSL dataset [1].

KW - Informatics

KW - sign language recognition

KW - top-down attention

KW - deep learning

UR - http://www.scopus.com/inward/record.url?scp=85180742060&partnerID=8YFLogxK

UR - https://www.mendeley.com/catalogue/6fa5f221-4e0f-376d-967b-385f6ae998c5/

U2 - 10.1109/icip49359.2023.10222729

DO - 10.1109/icip49359.2023.10222729

M3 - Article in conference proceedings

SN - 978-1-7281-9836-1

T3 - Proceedings - International Conference on Image Processing, ICIP

SP - 2555

EP - 2559

BT - 2023 IEEE International Conference on Image Processing, ICIP 2023 - Proceedings

PB - IEEE Electromagnetic Compatibility Society

CY - Piscataway

T2 - 2023 IEEE International Conference on Image Processing

Y2 - 8 October 2023 through 11 October 2023

ER -

DOI