Results 121 to 130 of about 14,068,918 (254)
PD-CBDM: Training Class-Balancing Diffusion Models with Perceptual Distinguish Loss
MathematicsFor image generation, denoising diffusion probabilistic models (DDPMs) have shown strong performance. Nevertheless, under class-imbalanced training data, many existing models tend to overfit head classes, which degrades image quality for tail classes. To
Junyan Hu +4 more
doaj +1 more source
Optical Generative Models for Novel Image Generation
We introduce optical generative models that transform digitally encoded optical seeds into novel images using a diffractive decoder. Experiments demonstrate monochrome and multi-color image generation, establishing a scalable and energy-efficient ...
Yuhang Li +4 more
core +1 more source
A context‐dependent modulatory role for eIF6 in acquired resistance to vemurafenib in melanoma
FEBS Letters, EarlyView.Acquired resistance to vemurafenib upregulates the translation factor eIF6 in melanoma cells. Silencing eIF6 in resistant cells reduces proliferation and partially restores drug sensitivity, whereas its overexpression increases sensitivity across melanoma lines regardless of BRAF status, via modulation of mTOR, S6K, and MAPK signaling.
George Kyriakopoulos +9 more
wiley +1 more source
Text as a Tool. The Effects of Using Image-Generating AI in German University Contexts
Journal of Digital PedagogyThis article investigates how German university students learn to use image-generating artificial intelligence and its implications for educational practice.
Maja Jerrentrup
doaj +1 more source
Metastatic niche shaped by host factors influences disseminated cancer cell fate
FEBS Letters, EarlyView.Metastasis is shaped not only by cancer cells but also by the environments they encounter. This review explores how factors such as aging, diet, the microbiome, lifestyle, and environmental exposures remodel organ‐specific niches in the lung, liver, bone, and brain, influencing where metastatic cells survive, remain dormant, or grow, and ultimately ...
Gwennan Delyth Ward +2 more
wiley +1 more source
Visual object networks: Image generation with disentangled 3D representation
, 2019 © 2018 Curran Associates Inc..All rights reserved. Recent progress in deep generative models has led to tremendous breakthroughs in image generation. However, while existing models can synthesize photorealistic images, they lack an understanding of our ...
Wu, Jiajun +6 more
core
FEBS Letters, EarlyView.Transient oligomers formed by intrinsically disordered proteins may be ‘invisible’ to direct detection yet remain accessible to solution NMR through equilibrium‐exchange measurements and pressure‐jump experiments. Complementary methods report on mass, stoichiometry, selected distance distributions, morphology, and internal packing.
Martin D. Gelenter, Ad Bax
wiley +1 more source
IEEE AccessWiFi Channel State Information (CSI)-based human image generation offers privacy-preserving sensing, but existing methods trained in one physical environment degrade sharply in unseen rooms. We attribute this to multipath propagation: static, environment-
Suan Lee, Dongmin Jin, Namhyeon Kim
doaj +1 more source
Neutralization of the PTX3/CD44 axis suppresses TGF‐β1 signaling in kidney fibrosis
FEBS Letters, EarlyView.This study investigates how the PTX3/CD44 axis promotes kidney fibrosis by activating TGF‐β1 signaling. We demonstrate that the neutralizing antibody WHC‐001 effectively inhibits this pathway, suppressing myofibroblast activation and partial epithelial–mesenchymal transition (EMT).
Chao‐Chun Cheng +8 more
wiley +1 more source
Product Image Generation Method Based on Morphological Optimization and Image Style Transfer
Applied SciencesIn order to improve the controllability and esthetics of product image generation, from the perspective of design, this study proposes a product image generation method based on morphological optimization, esthetic evaluation, and style transfer. Firstly,
Aimin Zhou +5 more
doaj +1 more source