Zobrazeno 1 - 10
of 5 707
pro vyhledávání: '"Salameh P."'
One key challenge to video restoration is to model the transition dynamics of video frames governed by motion. In this work, we propose TURTLE to learn the truncated causal history model for efficient and high-performing video restoration. Unlike tra
Externí odkaz:
http://arxiv.org/abs/2410.03936
Autor:
Jiang, Liyao, Hassanpour, Negar, Salameh, Mohammad, Singamsetti, Mohan Sai, Sun, Fengyu, Lu, Wei, Niu, Di
Text-to-image (T2I) diffusion models have demonstrated impressive capabilities in generating high-quality images given a text prompt. However, ensuring the prompt-image alignment remains a considerable challenge, i.e., generating images that faithful
Externí odkaz:
http://arxiv.org/abs/2408.11706
Autor:
Samadi, Mohammadreza, Han, Fred X., Salameh, Mohammad, Wu, Hao, Sun, Fengyu, Zhou, Chunhua, Niu, Di
Diffusion models have demonstrated strong performance in generative tasks, making them ideal candidates for image editing. Recent studies highlight their ability to apply desired edits effectively by following textual instructions, yet two key challe
Externí odkaz:
http://arxiv.org/abs/2408.08495
Autor:
Gusarov, Nikolay, Mandal, Rajesh, Salameh, Issa, Holzman, Itamar, Kvatinsky, Shahar, Ivry, Yachin
The rapid-pace growing demand for high-performance computation and big-data manipulation entails substantial increase in global power consumption, and challenging thermal management. Thus, there is a need in allocating competitive alternatives for co
Externí odkaz:
http://arxiv.org/abs/2405.18309
This paper addresses the challenge of learning to recite the Quran for non-Arabic speakers. We explore the possibility of crowdsourcing a carefully annotated Quranic dataset, on top of which AI models can be built to simplify the learning process. In
Externí odkaz:
http://arxiv.org/abs/2405.02675
In the field of robotics and computer vision, efficient and accurate semantic mapping remains a significant challenge due to the growing demand for intelligent machines that can comprehend and interact with complex environments. Conventional panoptic
Externí odkaz:
http://arxiv.org/abs/2405.02162
Autonomous vehicles often make complex decisions via machine learning-based predictive models applied to collected sensor data. While this combination of methods provides a foundation for real-time actions, self-driving behavior primarily remains opa
Externí odkaz:
http://arxiv.org/abs/2404.07383
Autor:
Mills, Keith G., Han, Fred X., Salameh, Mohammad, Lu, Shengyao, Zhou, Chunhua, He, Jiao, Sun, Fengyu, Niu, Di
Neural Architecture Search is a costly practice. The fact that a search space can span a vast number of design choices with each architecture evaluation taking nontrivial overhead makes it hard for an algorithm to sufficiently explore candidate netwo
Externí odkaz:
http://arxiv.org/abs/2403.13293
The end-to-end learning pipeline is gradually creating a paradigm shift in the ongoing development of highly autonomous vehicles, largely due to advances in deep learning, the availability of large-scale training datasets, and improvements in integra
Externí odkaz:
http://arxiv.org/abs/2403.12176
Autor:
Ghasemabadi, Amirhosein, Janjua, Muhammad Kamran, Salameh, Mohammad, Zhou, Chunhua, Sun, Fengyu, Niu, Di
Image restoration tasks traditionally rely on convolutional neural networks. However, given the local nature of the convolutional operator, they struggle to capture global information. The promise of attention mechanisms in Transformers is to circumv
Externí odkaz:
http://arxiv.org/abs/2401.15235