發(fā)布時(shí)間：2018-07-12 14:10

  本文選題：人臉對齊 + 人臉關(guān)鍵點(diǎn)定位��； 參考：《浙江大學(xué)》2017年碩士論文

【摘要】：人臉對齊是計(jì)算機(jī)視覺中的經(jīng)典問題之一,其目的是自動計(jì)算出給定人臉圖像中的面部關(guān)鍵點(diǎn)坐標(biāo)。精確的人臉關(guān)鍵點(diǎn)定位結(jié)果對許多視覺任務(wù)具有重要意義,如人臉識別、3D人臉重建、人臉表情分析、人臉姿態(tài)估計(jì)等。隨著相關(guān)技術(shù)的發(fā)展,目前的人臉對齊方法在受控條件下可以達(dá)到較低的定位誤差。然而,許多人臉相關(guān)應(yīng)用的輸入是在自然條件下獲取的,由于存在光照、背景、人臉姿態(tài)、圖像質(zhì)量等多種干擾因素,人臉對齊問題依然非常具有挑戰(zhàn)性。本文主要關(guān)注非受限條件下的人臉對齊問題,主要貢獻(xiàn)點(diǎn)如下:(1)本文通過實(shí)驗(yàn)分析發(fā)現(xiàn),合理的初始值可以使級聯(lián)回歸模型的定位誤差率大幅下降�；谠摪l(fā)現(xiàn),本文提出了由粗到精的兩階段人臉對齊算法框架,將人臉對齊分成粗定位和精定位兩個(gè)子問題,且每個(gè)問題應(yīng)該使用專用方法解決。(2)針對粗定位問題,本文設(shè)計(jì)并實(shí)現(xiàn)了一種基于深度卷積神經(jīng)網(wǎng)絡(luò)的模型,該模型以整張人臉為輸入,直接預(yù)測所有人臉關(guān)鍵點(diǎn)的位置坐標(biāo)。在300-W測試集上的結(jié)果表明,該模型能有效降低人臉關(guān)鍵點(diǎn)定位失敗率。(3)針對精定位問題,本文提出了一種基于參數(shù)共享的級聯(lián)回歸模型,該模型中的每個(gè)回歸步驟均使用相同的參數(shù)。與粗定位模型結(jié)合后,在300-W測試集上誤差率降低到了 state-of-the-art。此外,本文還指出可以將單個(gè)回歸模型作為梯度預(yù)測模型,通過結(jié)合梯度下降算法中的技巧,實(shí)驗(yàn)表明關(guān)鍵點(diǎn)定位誤差率還可以獲得進(jìn)一步下降。
[Abstract]:Face alignment is one of the classical problems in computer vision. Its purpose is to automatically calculate the coordinates of the key points in a given face image. Accurate location of key points is of great significance to many visual tasks, such as 3D face reconstruction, facial expression analysis, face pose estimation and so on. With the development of related technology, the current human face alignment method can achieve low positioning error under controlled conditions. However, the input of many human face related applications is obtained under natural conditions. Due to the existence of illumination, background, face pose, image quality and other interference factors, the problem of face alignment is still very challenging. The main contributions of this paper are as follows: (1) through experimental analysis, it is found that reasonable initial values can significantly reduce the localization error rate of cascaded regression models. Based on this discovery, a two-stage face alignment algorithm from coarse to fine is proposed in this paper. Face alignment is divided into two sub-problems: coarse location and fine location, and each problem should be solved by a special method. (2) aiming at the rough location problem, A model based on deep convolution neural network is designed and implemented in this paper. The model takes the whole face as input and directly predicts the position coordinates of the key points of all faces. The results on 300-W test set show that the model can effectively reduce the failure rate of face location at key points. (3) aiming at the problem of precise localization, a cascaded regression model based on parameter sharing is proposed in this paper. Each regression step in the model uses the same parameters. Combined with the rough localization model, the error rate on the 300-W test set is reduced to state-of-the-art. In addition, it is pointed out that a single regression model can be used as a gradient prediction model. By combining the techniques of gradient descent algorithm, the experiments show that the error rate of the key point location can be further reduced.
【學(xué)位授予單位】：浙江大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2017
【分類號】：TP391.41

【相似文獻(xiàn)】

相關(guān)碩士學(xué)位論文 前1條

1 王峰;基于兩階段定位模型的人臉對齊算法研究[D];浙江大學(xué);2017年

，

本文編號：2117411