Search Keywords

Results Restricted To:

https://www.temple.edu

Total Results: 5,230

Personalized Mobile Targeting with User Engagement Stages: Combining a ...

https://www.fox.temple.edu/sites/fox/files/SHMM_isre.2018.0831.pdf

Abstract. Low engagement rates and high attrition rates have been formidable challenges to mobile apps and their long-term success, especially for those whose revenues derive mainly from in-app purchases. To date, little is known about how companies can sci-entifically detect user engagement stages and optimize corresponding personalized-targeting promotion strategies to improve business ...

Learning Pixel-wise Alignment for Unsupervised Image Stitching

https://cis.temple.edu/~latecki/Papers/ACM_MM2023.pdf

Unlike image registration, im-age stitching has limited overlapping regions and large parallax, lacking constraints for non-overlapping regions in an unsupervised framework. Consequently, existing image registration methods are unable to produce correct ofsets for non-overlapping regions, and fail to output the whole image stitching result.

ArrayPipe: Introducing Job-Array Pipeline Parallelism for High ...

https://cis.temple.edu/~jiewu/research/publications/Publication_files/m48122-zhao%20final.pdf

DAPPLE [27] adopts one-forward-one-backward (1F1B) schedule to reduce the memory footprint of activations but introduces memory im-balance between devices. BPipe [5] leverages high-speed in-terconnects to transfer intermediate data between GPUs during training, enabling all GPUs to utilize comparable amounts of memory.

HDCFN: Haze Distribution-aware Cross-modal Fusion Network for Infrared ...

https://cis.temple.edu/~jiewu/research/publications/Publication_files/MM_2025_UAV_CR.pdf

HDCFN features two key components: (i) an infrared-guided multiscale feature enhance-ment framework that integrates haze-resistant structural cues from infrared modality with visible features across coarse to fine, im-proving the recovery of small objects, and (ii) a haze distribution-aware cross-modal fusion module that adaptively prioritizes ...

Mindy Shi's Homepage at Temple University

https://cis.temple.edu/~mindyshi/

Candidate must have or be close to obtaining a Ph.D. in quantitative fields (include but not limited to Computer Science, Statistics, Mathematics, and Physics) and have demonstrated a high level of research productivity through publication in peer-reviewed conferences and journals. Experiences with machine learning or data privacy is required. As a team player working closely with cross ...

Unit #3b - Temple University

https://community.mis.temple.edu/mis5214sec005spring2021/files/2020/03/MIS5214_Unit8_CaseStudy2_Maersk.pdf

2017, July – System upgraded (4,000 new servers, 45,000 new PC’s, with 2,500 applications) and computer-based business processes restored

Blended, Hybrid, and Flipped Courses: What’s the Difference?

https://sites.temple.edu/edvice/2019/11/05/blended-hybrid-and-flipped-courses-whats-the-difference/

Ariel Siegelman, Senior Instructional Technology Specialist If you’ve read about or attended workshops on approaches to teaching and learning with technology, chances are you’ve come across a few different terms to describe classes that have an online component. What are blended, hybrid, and flipped courses? Are they all describing the same approach to teaching, or are they different from ...

Lednet: A Lightweight Encoder-Decoder Network for Real-Time Semantic ...

https://cis.temple.edu/~latecki/Papers/ICIP2019.pdf

ABSTRACT The extensive computational burden limits the usage of CNNs in mobile devices for dense estimation tasks. In this paper, we present a lightweight network to address this prob-lem, namely LEDNet, which employs an asymmetric encoder-decoder architecture for the task of real-time semantic seg-mentation. More specifically, the encoder adopts a ResNet as backbone network, where two new ...

Microsoft Word - Skyscan_1172_MicroCT_Introduction[1]

https://medicine.temple.edu/sites/medicine/files/files/skyscan_1172_introduction.pdf

THE SKYSCAN 1172 MICROCT INTRODUCTION Principles of operation: Briefly, the system obtains multiple x-ray “shadow” transmission images of the object from multiple angular views as the object rotates on a high-precision stage. From these shadow images, cross-section images of the object are reconstructed using a modified Feldkamp cone-beam algorithm, creating a complete 3D representation of ...

An Overview of Cryptography - Temple University

https://cis.temple.edu/~qzeng/cis4360-spring17/papers/An%20Overview%20of%20Cryptography.pdf

For serious attackers with money to spend, such as some large companies or governments, Field Programmable Gate Array (FPGA) or Application-Specific Integrated Circuits (ASIC) technology offered the ability to build specialized chips that could provide even faster and cheaper solutions than a PC.