[LS2N_IPI] IRCCyN_IVC_Eyetracker_2008_11

Overview

The IRCCyN IVC Eyetracker SD 2008 11 video database contains 51 videos in SD format (720x576 pixels). There are 30 videos in interlaced at 50Hz (i50) and 21 videos in progressive format at 30Hz (p30).

All videos are H.264 coded with a bitrate to produce good quality. For the "i50" videos, there are 15 video contents with 2 versions for each content. One without and one with simulated transmission errors. For the "p30" videos, there are 16 video contents without transmission error and 5 of these contents with simulated transmission error.

It’s a free task experiment.

The videos and eyetracker raw data are provided.

This database is provided free of charge

Goal

Initial goal

This database was created to collect saliency maps to test a region of interest protectionsoftware on these videos.More details can be found in the related publications.A video quality database is available come from the results of this experiment to evaluate the quality impact of the protected region of interest.

Possible usage

This eye tracking SD video database can be used to validate a visual attention model takingcare about transmission error (simulated packets loss).This database can be used to evaluate the impact of transmission error on the saliency maps

Citation

Please, cite the following paper in your publications if you use this database in your work :Fadi Boulos, Wei Chen, Benoit Parrein and Patrick Le Callet, "Region-of-Interest Intra Prediction for H.264/AVC Error Resilience", IEEE ICIP, 2009.

## 概述（Overview）
IRCCyN IVC眼动追踪SD 2008 11视频数据库包含51段标准清晰度（Standard Definition, SD）格式视频，分辨率为720×576像素。其中30段为50Hz隔行扫描格式（i50），21段为30Hz逐行扫描格式（p30）。
所有视频均采用H.264编码，码率设置以保障优质画质。针对i50格式视频，共包含15类视频内容，每类内容均有两个版本：其一无传输误差模拟，其二带有传输误差模拟。针对p30格式视频，包含16类无传输误差的视频内容，其中5类额外带有传输误差模拟。
本数据集源自一项自由任务式眼动实验，提供原始视频与眼动追踪原始数据，且免费对外开放。

## 目标（Goal）
### 初始构建目标（Initial goal）
本数据集旨在采集显著性图（saliency map），以在上述视频上测试感兴趣区域保护软件。更多细节可参阅相关研究文献。基于本实验的结果可构建视频质量数据库，用于评估受保护感兴趣区域对视频质量的影响。

## 潜在应用场景（Possible usage）
本眼动追踪SD视频数据库可用于验证考虑传输误差（模拟数据包丢包）的视觉注意力模型；亦可用于评估传输误差对显著性图的影响。

## 引用要求（Citation）
若您在研究工作中使用本数据库，请在发表成果时引用以下论文：Fadi Boulos、Wei Chen、Benoit Parrein与Patrick Le Callet，《面向H.264/AVC错误鲁棒性的感兴趣区域帧内预测》，IEEE国际图像处理会议（IEEE ICIP），2009年。