Marco Raggio

Background estimation with Gaussian distribution for image segmentation, a fast approach

Adaptive background updating is one of the methods used to detect moving objects in video sequences. Many techniques have been presented in this field, but there are few mentions of the use of these methods in real-time applications. We concentrate on the speed of the algorithm and present a method that is fast enough to be used in video surveillance systems. We started from ideas presented using Gaussian distribution for background generation. Instead of using all the pixels in the image actively, we divide the pixels into active and inactive ones. Gaussian distributions are used to model the history of active pixels and to state whether they belong to background or foreground. According to the classification of the previous active pixel, the inactive pixels are also classified as a part of the background or foreground. We also reduce the frame frequency and use only every nth frame in the image sequence to construct the adaptive background. Some of the previous work and their results are first introduced. We then describe the method used by C. Stauffer and W.E.L. Grimson (see IEEE Conf. on Computer Vision & Pattern Recognition, p.246-52, 1999) and present our new ideas. The results are explained.

Search window size decision for motion estimation algorithm in H.264 video coder

In this paper, we propose, an innovative algorithm for reducing the complexity of the motion estimation (ME) module used in standard video compression applications. The idea is based on applying a motion detection algorithm for the decision of search window sizes in the motion estimation. The motion detection algorithm is based on blob coloring [D.H. Ballard and C. Brown, 1982] over sub-sampled binary images generated by block wise threshold image difference. This solution decreases the encoder complexity thanks to the reduction of SAD (sum of absolute difference) calculations especially in low complexity sequences like video-surveillance and video-telephony ones. The selected video compression algorithm for validating the proposed approach is H.264 [October 2002] that represent the state of the art of video codec.

An efficient multi-core SIMD implementation for H.264/AVC encoder

The optimization process of a H.264/AVC encoder on three different architectures is presented. The architectures are multiand singlecore and SIMD instruction sets have different vector registers size. The need of code optimization is fundamental when addressing HD resolutions with real-time constraints. The encoder is subdivided in functional modules in order to better understand where the optimization is a key factor and to evaluate in details the performance improvement. Common issues in both partitioning a video encoder into parallel architectures and SIMD optimization are described, and author solutions are presented for all the architectures. Besides showing efficient video encoder implementations, one of the main purposes of this paper is to discuss how the characteristics of different architectures and different set of SIMD instructions can impact on the target application performance. Results about the achieved speedup are provided in order to compare the different implementations and evaluate the more suitable solutions for present and next generation video-coding algorithms.

Multimedia-application-driven instruction set architecture simulation

This paper presents an application-driven architecture-design approach. VLIW architectures and instruction set simulation were chosen to fulfill multimedia domain requirements and to implement an efficient Hw-Sw co-design tool. Innovative approaches such as pipeline status modeling, simulation cache, and simulation oriented Hw description have been described. The performance of simulation tests for two validation case studies (TI TMS320C62x and ST200) are reported.

Real-time transmission and storage of video, audio, and health data in emergency and home care situations

The increase in the availability of bandwidth for wireless links, network integration, and the computational power on fixed and mobile platforms at affordable costs allows nowadays for the handling of audio and video data, their quality making them suitable for medical application. These information streams can support both continuous monitoring and emergency situations. According to this scenario, the authors have developed and implemented the mobile communication system which is described in this paper. The system is based on ITU-T H.323 multimedia terminal recommendation, suitable for real-time data/video/audio and telemedical applications. The audio and video codecs, respectively, H.264 and G723.1, were implemented and optimized in order to obtain high performance on the system target processors. Offline media streaming storage and retrieval functionalities were supported by integrating a relational database in the hospital central system. The system is based on low-cost consumer technologies such as general packet radio service (GPRS) and wireless local area network (WLAN or WiFi) for lowband data/video transmission. Implementation and testing were carried out for medical emergency and telemedicine application. In this paper, the emergency case study is described.

Health parameters and video live transmission with data storage on database, in emergency telemedicine

Mobile communication technologies could strongly help in facing health emergency situation. Thanks to wireless communication techniques, real time audio, video, medical parameters transmission and information storage are nowadays possible. Electrocardiogram, blood pressure and other patients information are sent to a remote hospital/medical database center server. Actions taken by the emergency staff are stored for real time operative-center interaction and for a successive database consultation. The emergency video recording is often useful as legal evidence in case of disputation. In this paper a real time connection telemedical system is described. The system is based on low-cost consumer technologies such as GPRS and WiFi for low-band data/video transmission, H.264 for the real time video stream encoding and database video storage. A relational database has been used for store and query data All system is based on Microsoft .Net technology allowing great portability and stability to overall modules, such as HMI (human machine interface) and multimedia codecs

An optimized software implementation of the HEVC/H.265 video decoder

In this paper an optimized implementation of the HEVC video decoder is shown. The solutions developed to support the new features of HEVC are shown together with the achieved performance. The HEVC decoder complexity has been evaluated and the most demanding modules have been optimized exploiting SIMD instructions. Even though the here described concepts have a general value, the effectiveness of the proposed solutions has been verified on the ARM architecture. The selected architecture is ARM Cortex A9 with NEON SIMD extension. We will demonstrate that the resulting real-time HEVC software decoder can decode 720p (1280x720) streams at 30 frames per second on a single core ARMv7 at 1.2 GHz.

A Simulation and Exploration Technology for Multimedia-Application-Driven Architectures

The increasing of computational power requirements for DSP and Multimedia application and the needs of easy-to-program development environment has driven recent programmable devices toward Very Long Instruction Word (VLIW) [1] architectures and Hw-Sw co-design environments [2]. VLIW architecture allows generating optimized machine code from high-level languages exploiting Instruction Level Parallelism (ILP) [3]. Furthermore, applications requirements and time to market constraints are growing dramatically moving functionalities toward System on Chip (SoC) direction. This paper presents VLIW-SIM, an Application-Driven Architecture-design approach based on Instruction Set simulation. VLIW architectures and Instruction Set simulation were chosen to fulfill multimedia domain requirements and to implement an efficient Hw-Sw co-design environment. The VLIW-SIM simulation technology is based on pipeline status modeling, Simulation cache and Simulation Oriented Hw description. An effective support for Hw-Sw co-design requires high simulation performance (in terms of Simulated Instruction per Second—SIPS), flexibility (the ability to represent a number of different architectures) and cycle accuracy. There is a strong trade-off between these features: cycle accurate or close to cycle accurate simulation have usually low performance [4, 5]. Good simulation performance can be obtained loosing the simulator flexibility. Moreover SoC simulation requires a further degree of flexibility in simulating different components (core, co-processors, memories, buses). The proposed approach is focused on interpretative (not compiled [6]) re-configurable Instruction Set Simulator (ISS) in order to support both application design and architecture exploration. VLIW-SIM main features are: efficient host resource allocation, Instruction Set and Architecture description Flexibility (Instruction Set Dynamic Generation and Simulation Oriented Hardware Description), Step by step pipeline status tracking, Simulation Speed and Accuracy. Performance of simulation test for three validation case studies (TI TMS320C62x, TI TMS320C64x and ST200) are reported.

Real time scalable video communication demonstration at ICMCS-99

This document describes the demonstration of the Scaleable Video Communication Unit (VCU) at ICMCS-99 in Florence. The demonstration simulates two different field-applications: Distance Learning (DL) and Forest Surveillance (FS). The VCU is based on a scalable H263+ video codec integrated in a H324 multimedia terminal used as a common core for the two applications addressed in the SCALAR field tests.

Supervision analysis and control system of photovoltaic power plants

This paper describes a supervision system able to handling the data collection from photovoltaic implants, their analysis allowing providing prevision and control of energy production. The supervision system integrates different functionality usually demanded to single subsystems and advanced features devoted to the implants maintenance and security. The supervision system is able to interact with the PV plant varying its functionality.