Yan se shi jiao que xian zhe yu zheng chang shi jiao ren shi jin xing tu pian yu shi pin de shi jiao gong xiang

Ph.D.The two distinctive visual experiences of binocular display, with and without stereoscopic glasses, have been recently utilized for visual sharing between the normal-vision audiences and the audiences suffering from color vision deficiency (sometimes called color blindness or color weakness). However, the visual sharing problem is still challenging due to the need of simultaneous satisfaction of multiple binocular visual requirements, in order to offer a color-distinguishable and binocularly fusible visual experience to audiences suffering from color vision deficiencies (CVD), without hurting the visual experience of the normal-vision audiences. Existing methods are not quite suitable for practical usage. The first reason is that their optimization is unstable and time-consuming in nature. In our first application, we propose a deep-learning based solution for solving this problem. Our method outperforms the existing solution in terms of all evaluation metrics. To achieve this, we propose to formulate this binocular image generation problem as a generation problem of a difference image, which can effectively enforce the binocular constraints. We also propose to retain only high-quality training data and enrich the variety of training data via intentionally synthesizing various confusing color combinations. With these, we train up a high-quality neural network model. Through multiple quantitative measurements and user study, we demonstrate this learning-based approach can significantly improve the quality of generated results with fast performance.Besides, existing methods only work for still images, and lack of temporal consistency for video application. In this thesis, we also introduce two works on synthesizing colorblind-sharable videos that possess the temporal consistency for both visual experiences of colorblind and normal-vision, and retains all other crucial characteristics for visual sharing with colorblind. In one work, we formulate this challenging multi-constraint problem as a global optimization problem and minimize an objective function consisting of temporal term, color preservation term, color distinguishability term, and binocular fusibility term. In the other work, we train a neural network to predict the temporal coherent polynomial coefficients in the global color decomposed domain without introducing color inconsistency (same color regions map to different colors) as in the former one. Qualitative and quantitative experiments are conducted to evaluate the effectiveness of the two proposed method comparing to existing methods.雙目顯示器提供戴與不戴眼鏡這兩種不同的視覺體驗，它們最近被用到了顏色視覺缺陷者(有時也稱色盲或者色弱者)與正常視覺人士的視覺共享中。然而，由於要同時滿足多個雙目視覺要求以在不影響正常人視覺的前提下為顏色視覺缺陷者提供顏色可分和雙目可融合的視覺體驗，这个问题仍然是有挑战性的。當前的方法由於不穩定性和過多的時間消耗，導致它們不適合用於實際中。在我們第一個應用中，我們提出了第一個基於深度學習的視覺共享解決方案。我們的方法在所有的評價指標中都優於之前的方法。為了做到這點，我們將雙目圖片生成問題公式化為差距圖片的生成問題。這樣我們可以有效地保證雙目約束。我們也提出只保留高質量的訓練數據，並且通過有目的性地生成多種難以區分的顏色組合來使訓練數據更加豐富。通過這些手段，我們訓練了一個高質量的神經網絡模型。通過多種定量評估和用戶實驗，我們展現了這個基於學習的方法可以快速生成更高質量的結果。除此之外，當前的方法只能處理單張圖片，缺乏可用於視頻應用的時序一致性。在這篇論文中，我們還介紹了兩個生成可與顏色視覺缺陷者共享的視頻的工作，生成的視頻具有對於顏色視覺缺陷者和正常視覺人士的視覺體驗的時序一致性，並且保持所有為了進行視覺共享的其他關鍵特性。在其中一個工作中，我們將這個有挑戰性的多制約問題公式化為一個全局最優化問題，並且最小化一個包含了時序項，顏色保持項，顏色區分項和雙目融合項的目標方程。在另一個工作中，我們訓練了一個神經網絡來全局顏色分解的時序一致性的多項式係數，從而不會引入在前一個工作中可能出現的顏色不一致性（同樣的顏色區域被轉換為不同的顏色）。我們進行了定性和定量分析來驗證這兩個方法相對于其他已有方法的有效性。Hu, Xinghong.Thesis Ph.D. Chinese University of Hong Kong 2019.Includes bibliographical references (leaves 134-145).Abstracts also in Chinese.Title from PDF title page (viewed on 20, May 2021).Hu, Xinghong