Zobrazeno 1 - 10
of 495
pro vyhledávání: '"I.4.5"'
We propose an effective method for inserting adapters into text-to-image foundation models, which enables the execution of complex downstream tasks while preserving the generalization ability of the base model. The core idea of this method is to opti
Externí odkaz:
http://arxiv.org/abs/2410.22901
Four-dimensional scanning transmission electron microscopy (4D-STEM) is a powerful tool that allows for the simultaneous acquisition of spatial and diffraction information, driven by recent advancements in direct electron detector technology. Althoug
Externí odkaz:
http://arxiv.org/abs/2410.17206
Autor:
Barua, Hrishav Bakul, Kalin, Stefanov, Che, Lemuel Lai En, Abhinav, Dhall, KokSheik, Wong, Ganesh, Krishnasamy
Low Dynamic Range (LDR) to High Dynamic Range (HDR) image translation is an important computer vision problem. There is a significant amount of research utilizing both conventional non-learning methods and modern data-driven approaches, focusing on u
Externí odkaz:
http://arxiv.org/abs/2410.15068
We propose a novel neural network approach, LARP (Learned Articulated Rigid body Physics), to model the dynamics of articulated human motion with contact. Our goal is to develop a faster and more convenient methodological alternative to traditional p
Externí odkaz:
http://arxiv.org/abs/2410.12023
Sonar image synthesis is crucial for advancing applications in underwater exploration, marine biology, and defence. Traditional methods often rely on extensive and costly data collection using sonar sensors, jeopardizing data quality and diversity. T
Externí odkaz:
http://arxiv.org/abs/2410.08612
Autor:
Fang, Irving, Shi, Kairui, He, Xujin, Tan, Siqi, Wang, Yifan, Zhao, Hanwen, Huang, Hung-Jui, Yuan, Wenzhen, Feng, Chen, Zhang, Jing
Humans effortlessly integrate common-sense knowledge with sensory input from vision and touch to understand their surroundings. Emulating this capability, we introduce FusionSense, a novel 3D reconstruction framework that enables robots to fuse prior
Externí odkaz:
http://arxiv.org/abs/2410.08282
We propose two new evaluation metrics to assess realness of generated images based on normalizing flows: a simpler and efficient flow-based likelihood distance (FLD) and a more exact dual-flow based likelihood distance (D-FLD). Because normalizing fl
Externí odkaz:
http://arxiv.org/abs/2410.02004
Inconsistent responses of X-ray detector elements lead to stripe artifacts in the sinogram data, which manifest as ring artifacts in the reconstructed CT images, severely degrading image quality. This paper proposes a method for correcting stripe art
Externí odkaz:
http://arxiv.org/abs/2409.15731
Previous studies have shown that faces are rated as more attractive when they are partially occluded. The cause of this observation remains unclear. One explanation is a mental reconstruction of the occluded face parts which is biased towards a more
Externí odkaz:
http://arxiv.org/abs/2409.15443
Autor:
Fischer, Yann, Auray, Zacharie, Martinache, Olivier, Dubosc, Marius, Topéza, Noé, Magnier, Chloé, Boy-Arnould, Maxime, Atlan, Michael
This study presents a novel quantitative estimation method for total retinal arterial blood flow utilizing real-time Doppler holography at an unprecedented frame rate of 33,000 frames per second. This technique, leveraging high-speed digital holograp
Externí odkaz:
http://arxiv.org/abs/2409.17180