Idioma:

Visual Question Answering Models for Zero-Shot Pedestrian Attribute Recognition: A Comparative Study

Castrillón-Santana, Modesto ; Sánchez-Nielsen, Elena ; Freire-Obregón, David ; Santana, Oliverio J. ; Hernández-Sosa, Daniel ; Lorenzo-Navarro, Javier

SN computer science, 2024-06, Vol.5 (6), p.680, Article 680 [Periódico revisado por pares]

Singapore: Springer Nature Singapore

Texto completo disponível

Citações Citado por

Enviar para

Título:
Visual Question Answering Models for Zero-Shot Pedestrian Attribute Recognition: A Comparative Study
Autor: Castrillón-Santana, Modesto ; Sánchez-Nielsen, Elena ; Freire-Obregón, David ; Santana, Oliverio J. ; Hernández-Sosa, Daniel ; Lorenzo-Navarro, Javier
Assuntos: Annotations ; Artificial intelligence ; Biometrics ; Comparative studies ; Competition ; Computer Analysis of Images and Patterns in the Deep Learning Era ; Computer Imaging ; Computer Science ; Computer Systems Organization and Communication Networks ; Data Structures and Information Theory ; Datasets ; Gender ; Information Systems and Communication Service ; Language ; Large language models ; Natural language ; Original Research ; Pattern Recognition and Graphics ; Questions ; Recognition ; Semantics ; Software Engineering/Programming and Operating Systems ; State-of-the-art reviews ; Surveillance ; Vision
É parte de: SN computer science, 2024-06, Vol.5 (6), p.680, Article 680
Descrição: Pedestrian Attribute Recognition (PAR) poses a significant challenge in developing automatic systems that enhance visual surveillance and human interaction. In this study, we investigate using Visual Question Answering (VQA) models to address the zero-shot PAR problem. Inspired by the impressive results achieved by a zero-shot VQA strategy during the PAR Contest at the 20th International Conference on Computer Analysis of Images and Patterns in 2023, we conducted a comparative study across three state-of-the-art VQA models, two of them based on BLIP-2 and the third one based on the Plug-and-Play VQA framework. Our analysis focuses on performance, robustness, contextual question handling, processing time, and classification errors. Our findings demonstrate that both BLIP-2-based models are better suited for PAR, with nuances related to the adopted frozen Large Language Model. Specifically, the Open Pre-trained Transformers based model performs well in benchmark color estimation tasks, while FLANT5XL provides better results for the considered binary tasks. In summary, zero-shot PAR based on VQA models offers highly competitive results, with the advantage of avoiding training costs associated with multipurpose classifiers.
Editor: Singapore: Springer Nature Singapore
Idioma: Inglês

Voltar para lista de resultados

Anterior Resultado 10 Avançar Ir para próxima página

Realização: Logos de Redes Sociais:

Visual Question Answering Models for Zero-Shot Pedestrian Attribute Recognition: A Comparative Study

Castrillón-Santana, Modesto ; Sánchez-Nielsen, Elena ; Freire-Obregón, David ; Santana, Oliverio J. ; Hernández-Sosa, Daniel ; Lorenzo-Navarro, Javier

Singapore: Springer Nature Singapore

Buscando em bases de dados remotas. Favor aguardar.