A5012374720- Parallel Computing and Optimization Techniques
- Embedded Systems Design Techniques
- Interconnection Networks and Systems
- Distributed and Parallel Computing Systems
- Electrowetting and Microfluidic Technologies
- Software Engineering Research
- Anatomy and Medical Technology
- Software System Performance and Reliability
- Advanced Data Storage Technologies
- Advanced Image and Video Retrieval Techniques
- Low-power high-performance VLSI design
- Computer Graphics and Visualization Techniques
- Music and Audio Processing
- Network Packet Processing and Optimization
- Augmented Reality Applications
- Video Analysis and Summarization
Friedrich-Alexander-Universität Erlangen-Nürnberg
2018-2023
International Audio Laboratories Erlangen
2021
It is well known that to accelerate stencil codes on CPUs or GPUs and exploit hardware caches their lines optimizers must find spatial temporal locality of array accesses harvest data-reuse opportunities. On FPGAs there the burden are no built-in (or only pre-built descriptions for cache blocks inefficient codes). But this paper demonstrates lack also a chance as polyhedral methods can be used generate stencil-specific cache-structures right sizes FPGA fill flush them efficiently with wide...
Flexible retrieval systems are required for conveniently browsing through large music collections. In a particular content-based scenario, the user provides query audio snippet, and system returns recordings from collection that similar to query. this fast response is essential positive experience. For realizing low times, one requires index structures facilitate efficient search operations. One such structure K-d tree, which has already been used in systems. As an alternative, we propose...