AI Article Synopsis
- Image style transfer allows the exchange of different visual styles between images, playing a crucial role in big data systems.
- A common issue in deep learning methods for style transfer is the loss of content information, which this paper addresses by incorporating L1 loss and perceptual loss into the VGG-19 network.
- The improved model successfully enhances style transfer capabilities while preserving content, with notable increases in structural similarity, cosine similarity, and mutual information by 0.323%, 0.094%, and 3.591%, respectively.
Article Abstract
Image style transfer can realize the mutual transfer between different styles of images and is an essential application for big data systems. The use of neural network-based image data mining technology can effectively mine the useful information in the image and improve the utilization rate of information. However, when using the deep learning method to transform the image style, the content information is often lost. To address this problem, this paper introduces L1 loss on the basis of the VGG-19 network to reduce the difference between image style and content and adds perceptual loss to calculate the semantic information of the feature map to improve the model's perceptual ability. Experiments show that the proposal in this paper improves the ability of style transfer, while maintaining image content information. The stylization of the improved model can better meet people's requirements for stylization, and the evaluation indexes of structural similarity, cosine similarity, and mutual information value have increased by 0.323%, 0.094%, and 3.591%, respectively.
Download full-text PDF |
Source |
|---|---|
| http://www.ncbi.nlm.nih.gov/pmc/articles/PMC8407993 | PMC |
| http://dx.doi.org/10.1155/2021/8387382 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Extracellular and intracellular digestion in the bivalves studied by magnetic resonance imaging with a contrast reagent.
J Exp Biol
January 2025
Division of Cell Structure, National Institute for Physiological Sciences, Okazaki, Aichi 444-8787, Japan.
We investigated the extracellular and intracellular digestion of bivalves employing magnetic resonance imaging (MRI). Ruditapes philippinarum clams and Mytilus galloprovincialis mussels were incubated in seawater containing a contrast reagent (GdDTPA) at 20°C. The digestive systems, from the esophagus to the rectum, were visualized at a high signal intensity by the T1-weighted MRI.
View Article and Find Full Text PDFIntravenous Tranexamic Acid Does Not Improve Visual Field Clarity During Hip Arthroscopy: A Double-Blind, Randomized, Placebo-Controlled Trial.
Arthroscopy
January 2025
Department of Orthopaedic Surgery, Hospital for Special Surgery, New York, NY, USA.
Purpose: To determine the effectiveness of administering intravenous (IV) tranexamic acid (TXA) on altering visual field clarity (VFC) during arthroscopic hip preservation surgery for patients with femoroacetabular impingement syndrome (FAIS).
Methods: This randomized, double-blind, parallel-design trial was conducted over a seven-month period between October 2023 and May 2024 at a single tertiary musculoskeletal hospital. Inclusion criteria included consecutive patients that were diagnosed with FAIS through clinical history, physical exam and advanced imaging and indicated for hip arthroscopy after having failed conservative management.
Advancing precision agriculture with deep learning enhanced SIS-YOLOv8 for Solanaceae crop monitoring.
Front Plant Sci
January 2025
College of Information Technology, Jilin Agricultural University, Changchun, China.
Introduction: Potatoes and tomatoes are important Solanaceae crops that require effective disease monitoring for optimal agricultural production. Traditional disease monitoring methods rely on manual visual inspection, which is inefficient and prone to subjective bias. The application of deep learning in image recognition has led to object detection models such as YOLO (You Only Look Once), which have shown high efficiency in disease identification.
View Article and Find Full Text PDFAdvancements in AI Medical Education: Assessing ChatGPT's Performance on USMLE-Style Questions Across Topics and Difficulty Levels.
Cureus
December 2024
Medical Education, University of South Florida Morsani College of Medicine, Tampa, USA.
Background AI language models have been shown to achieve a passing score on certain imageless diagnostic tests of the USMLE. However, they have failed certain specialty-specific examinations. This suggests there may be a difference in AI ability by medical topic or question difficulty.
View Article and Find Full Text PDFExposure to residential air pollution and the development of functional connectivity of brain networks throughout adolescence.
Environ Int
January 2025
ISGlobal, Barcelona, Spain; Universitat Pompeu Fabra, Barcelona, Spain; Department of Child and Adolescent Psychiatry/Psychology, Erasmus University Medical Centre, Rotterdam, the Netherlands; Spanish Consortium for Research on Epidemiology and Public Health (CIBERESP), Instituto de Salud Carlos III, Madrid, Spain; ICREA, Barcelona, Spain. Electronic address:
Background: A few studies linked air pollution to differences in functional connectivity of resting-state brain networks in children, but how air pollution exposure affects the development of brain networks remains poorly understood. Therefore, we studied the association of air pollution exposure from birth to 3 years and one year before the first imaging assessment with the development of functional connectivity across adolescence.
Methods: We utilized data from 3,626 children of the Generation R Study (The Netherlands).
Want AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!