本文選題：立體視頻 + 立體掩蔽效應(yīng)��； 參考：《光電子·激光》2017年09期

【摘要】：針對(duì)立體視頻中存在著大量的感知冗余,首先通過(guò)主觀實(shí)驗(yàn)研究梯度和紋理因素對(duì)立體掩蔽效應(yīng)的影響,建立雙目恰可察覺(jué)編碼失真(BJNCD)模型;同時(shí)考慮到人類(lèi)視覺(jué)特性(HVS)的視覺(jué)敏感度并非恒定不變,當(dāng)視網(wǎng)膜離心率變大時(shí),像素的視覺(jué)閾值也隨之變大,因此結(jié)合HVS的視網(wǎng)膜中心凹感知特性建立了基于中心凹的雙目恰可察覺(jué)編碼失真(FBJNCD)模型。為驗(yàn)證所提出模型的有效性,將其應(yīng)用于多視點(diǎn)高效視頻編碼(MV-HEVC)測(cè)試平臺(tái)中對(duì)立體視頻進(jìn)行非對(duì)稱(chēng)編碼。實(shí)驗(yàn)結(jié)果表明,所提出模型在保持立體視頻感知質(zhì)量的同時(shí)能夠有效節(jié)省編碼碼率,提高了立體視頻壓縮效率。
[Abstract]:There is a large amount of perceptual redundancy in stereo video. Firstly, the influence of gradient and texture factors on stereo masking effect is studied through subjective experiments, and the BJNCD-model of binocular perceptible coding distortion is established.Considering that the visual sensitivity of HVS is not constant, as the retinal eccentricity increases, the visual threshold of pixels increases.Therefore, a binocular exactly detectable coding distortion (FBJNCD) model based on central fovea is established in combination with HVS's fovea perception.In order to verify the validity of the proposed model, the proposed model is applied to asymmetric coding of stereo video in a multi-view efficient video coding test platform.Experimental results show that the proposed model can effectively save the coding rate and improve the stereo video compression efficiency while maintaining the perceptual quality of stereo video.
【作者單位】： 寧波大學(xué)信息科學(xué)與工程學(xué)院;南京大學(xué)計(jì)算機(jī)軟件新技術(shù)國(guó)家重點(diǎn)實(shí)驗(yàn)室;
【基金】：國(guó)家自然科學(xué)基金(U1301257,61271270,61671258) 國(guó)家高技術(shù)研究發(fā)展計(jì)劃(2015AA015901) 浙江省自然科學(xué)基金(LY15F010005)資助項(xiàng)目
【分類(lèi)號(hào)】：TN919.81
，

本文編號(hào)：1735111