4 sonuçlar
Arama Sonuçları
Listeleniyor 1 - 4 / 4
Yayın Real-time video frame differentiator based on DDR3 SDRAM memory interface(IEEE Computer Society, 2018-08) Davutoğlu, Doğancan; Yıldız, Nerhun; Tavşanoğlu, Ahmet Vedat; Ayten, Umut EnginIn this paper, design of a real-time video frame differentiator based on an external memory interface is proposed. Furthermore, implementation and simulation processes of the design is discussed. The proposed design is capable of differentiating video frames over time, up to full-HD resolution at 60 Hz frame rate. An external SDRAM memory unit is used within the proposed design and drived by a memory interface. In order to improve the flexibility of the architecture, video resolution, video buffer size on memory and burst size of the memory interface are designed to be user defined and configurable.Yayın On the way to a third generation real-time cellular neural network processor(IEEE Computer Society, 2016) Yıldız, Nerhun; Cesur, Evren; Tavşanoğlu, Ahmet VedatIn this proceeding, the architecture of a third generation Real-Time Cellular Neural Network (CNN) Processor (RTCNNP-v3) is disclosed, which is a digital CNN emulator to be implemented on an FPGA device. The previous generation emulator, RTCNNP-v2, is the only CNN implementation reported to be capable of processing full-HD 1080p@60 (1080×1920 resolution at 60 Hz frame rate) video images in real-time. However, there are some weaknesses in both the design and implementation of RTCNNP-v2, like the inability to process different parts of the video images in parallel, lack of support for recording and recalling intermediate frames using external memory and it has some jitter issues at computation rates above 200 MHz. All of those issues are addressed in the next architecture of our CNN emulator, RTCNNP-v3, which is being implemented of an FPGA device.Yayın Design of a third generation real-time cellular neural network emulator(IEEE, 2014) Yıldız, Nerhun; Cesur, Evren; Tavşanoğlu, Ahmet VedatIn this paper, the features of the next generation Real-Time Cellular Neural Network Processor (RTCNNP-v3) are discussed. The RTCNNP-v2 structure is the only CNN implementation that is reported to be capable of processing full-HD 1080p@60 (1920 x 1080 resolution at 60 Hz frame rate) video images in real-time, due to its fully-pipelined architecture, however, it has some weaknesses like the inability to divide the processing in spatial domain, record and recall intermediate results to an external memory and has some issues in its internal memory coding. Those shortcomings are to be addressed in the next design of our CNN emulator - RTCNNP-v3, which will increase the range of applications and enable the implementation to match the requirements of the cutting-edge movie production technologies like UHD (4K) and the future FUHD (8K).Yayın Demo: Real-time video frame differentiator based on external memory interface(IEEE Computer Society, 2018-08) Davutoğlu, Doğancan; Yıldız, Nerhun; Tavşanoğlu, Ahmet Vedat; Ayten, Umut EnginImplementation and demonstration processes of a real-time video frame differentiator based on an external memory interface is described in this paper. The video frame differentiation process is successfully implemented on both low cost and high-end FPGA development boards, then demonstrated by using sample videos at 1024x768@60 and 1920x1080@60 resolutions. Input video resolution, video buffer size on memory and burst size of the memory interface can be configured before implementation.
