MulticoreWare and Cadence Target Embedded Neural Network Applications

  • On May 2, 2016
MulticoreWare has partnered with Cadence to develop high-performance convolutional neural-network (CNN) applications on their latest Vision P5 and Vision P6 DSPs. “We have been working closely with Cadence to develop CNN-based vision applications. Features such as wide vector SIMD processing, VLIW instructions, and fast histogram and scatter/gather intrinsics make it an ideal platform for demanding CNN […]
Read More

MulticoreWare Announces UHDkit Advanced HEVC Video Encoding Library, powered by x265

  • On April 18, 2016
Developers of x265, the world’s leading HEVC / H.265 encoder, debut an extended library for faster, more efficient HEVC video encoding. SUNNYVALE, Calif., April 18, 2016 /PRNewswire/ — MulticoreWare, developers of the category-leading x265 HEVC video encoder, announce the immediate availability of UHDkit, an extended encoding library built on top of the x265 HEVC encoder. […]
Read More

MulticoreWare Announces Release of New MxPA OpenCL compiler for DSPs

  • On March 16, 2016
MulticoreWare Brings Unprecedented Performance and Productivity to New Processor Platforms (SUNNYVALE, CA) March 16, 2015 – In the highly competitive mobile and embedded device markets, manufacturers are increasingly developing sophisticated applications that rely on powerful image processing algorithms for gesture recognition, object recognition, 3D scanning and modeling, augmented reality, computational photography, etc. New processor architectures, […]
Read More

x265 rated best overall HEVC encoder

  • On October 26, 2015
The x265 HEVC encoder has been rated as the best overall encoder in the first HEVC codec comparison by the Moscow State University Graphics & Media Lab. Moscow State University has been studying video compression for many years, and their regular codec comparison reports are well known for being comprehensive and thorough investigations of video encoder […]
Read More

MulticoreWare has Developed High Performance Face Detection Neural Network for Embedded Vision Processors

  • On July 9, 2015
MulticoreWare, Inc. has developed a high performance face detection algorithm solution for Synopsys, Inc.’s DesignWare® EV Family of vision processors utilizing Convolutional Neural Networks (CNNs). The collaboration demonstrates that high levels of accuracy and performance can be achieved in object detection software algorithms on a highly optimized embedded hardware platform. Neural network object detection solutions […]
Read More

MulticoreWare joins Xilinx Alliance Program

  • On May 28, 2015
MulticoreWare, Inc. is pleased to join the Xilinx Alliance Program, a worldwide ecosystem of qualified companies collaborating with Xilinx Inc. to further the development of All Programmable technologies. MulticoreWare is also announcing the release of a Convolutional Neural Networks (CNN) based face detection tool for the Xilinx Virtex-7 FPGA, developed using the SDAccel development environment. […]
Read More