Conference
Improving Person Re-identification Through Low-Light Image Enhancement
| العنوان: | Improving Person Re-identification Through Low-Light Image Enhancement |
|---|---|
| المؤلفون: | Santana, Oliverio J., Lorenzo-Navarro, Javier, Freire-Obregón, David, Hernández-Sosa, Daniel, Castrillón-Santana, Modesto |
| المساهمون: | orcid:0000-0001-7511-5783, orcid:0000-0002-2834-2067, orcid:0000-0003-2378-4277, orcid:0000-0003-3022-7698, orcid:0000-0002-8673-2725, 7003605046, 15042453800, 23396618800, 6507124168, 57218418238, BU-INF |
| المصدر: | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) [ISSN 0302-9743], v. 14547 LNCS, p. 95-110, (Enero 2024) |
| سنة النشر: | 2024 |
| المجموعة: | Universidad de Las Palmas de Gran Canaria: Acceda |
| مصطلحات موضوعية: | 220990 Tratamiento digital. Imágenes, Computer Vision, In The Wild Dataset, Low-Light Image Enhancement, Person Re-Identification, Sporting Event, Ultra-Trail Running |
| الوصف: | Person re-identification (ReID) is a popular area of research in the field of computer vision. Despite the significant advancements achieved in recent years, most of the current methods rely on datasets containing subjects captured with good lighting under static conditions. ReID presents a significant challenge in real-world sporting scenarios, such as long-distance races that take place over varying lighting conditions, ranging from bright daylight to night-time. Unfortunately, increasing the exposure time on the capture devices to mitigate low-light environments is not a feasible solution, as it would result in blurry images due to the motion of the runners. This paper surveys several low-light image enhancement methods and finds that including an image pre-processing step in the ReID pipeline before extracting the distinctive body features of the subjects can lead to significant improvements in performance. ; 110 ; 95 |
| نوع الوثيقة: | conference object |
| اللغة: | English |
| ردمك: | 978-3-031-54725-6 978-85-18-76722-6 978-3-031-54726-3 3-031-54725-X 85-18-76722-9 3-031-54726-8 |
| تدمد: | 0302-9743 1611-3349 |
| Relation: | Interaccióny Re-Identificación de Personas Mediante Machine Learning, Deep Learningy Análisis de Datos Multimodal: Hacia Una Comunicación Más Natural en la Robótica Social; Re-identificación mUltimodal de participaNtes en competiciones dEpoRtivaS; Infraestructura de Computación Científica Para Aplicaciones de Inteligencia Artificialy Simulación Numérica en Medioambientey Gestión de Energías Renovables (Iusiani-Ods); La viabilidad jurídica del documento de voluntades anticipadas; Lecture Notes in Computer Science; 12th International Conference on Pattern Recognition Applications and Methods (ICPRAM 2023); 14547 LNCS; Scopus; http://hdl.handle.net/10553/129687; 85187672299; events152778; Sí |
| DOI: | 10.1007/978-3-031-54726-3_6 |
| الاتاحة: | http://hdl.handle.net/10553/129687 https://doi.org/10.1007/978-3-031-54726-3_6 |
| رقم الانضمام: | edsbas.9F310179 |
| قاعدة البيانات: | BASE |
| ردمك: | 9783031547256 9788518767226 9783031547263 303154725X 8518767229 3031547268 |
|---|---|
| تدمد: | 03029743 16113349 |
| DOI: | 10.1007/978-3-031-54726-3_6 |