SmartPaint:一種基于生成式對抗神經(jīng)網(wǎng)絡(luò)的人機協(xié)同繪畫系統(tǒng)(英文)
發(fā)布時間:2023-04-16 14:00
當前人工智能在模仿和大批量生產(chǎn)設(shè)計作品中扮演重要角色(如電商廣告),而在與用戶合作創(chuàng)作時表現(xiàn)欠佳。人們有能力使用草圖表達創(chuàng)意想法,但缺乏專業(yè)繪畫技巧完成精美畫作。已有人工智能方法無法基于用戶輸入草圖的語義輸出具有藝術(shù)美感的畫作。本文開發(fā)了一種基于生成式對抗神經(jīng)網(wǎng)絡(luò)的人機協(xié)作繪畫系統(tǒng)——SmartPaint,支持人機合作創(chuàng)作動漫風(fēng)景畫作。該系統(tǒng)使用動漫圖像數(shù)據(jù)及其相應(yīng)語義標注圖、邊緣檢測圖訓(xùn)練生成式對抗神經(jīng)網(wǎng)絡(luò)。通過此種方式,該系統(tǒng)能夠同時理解動漫風(fēng)格以及風(fēng)景圖像中物體的語義和空間關(guān)系。在使用中,用戶輸入草圖作為語義標注圖,系統(tǒng)自動為其合成邊緣圖;根據(jù)合成的邊緣圖生成具有恰當風(fēng)格紋理的畫作,從而穩(wěn)定地處理多樣化草圖。實驗證明該系統(tǒng)可有效滿足用戶創(chuàng)作需求,生成高質(zhì)量動漫風(fēng)格畫作。
【文章頁數(shù)】:14 頁
【文章目錄】:
1 Introduction
2 Related work
2.1 Understanding user inputs
2.2 Turning a sketch into a painting
3 System overview and implementation
3.1 System overview
1.Painting producer
2.Edge synthesizer
3.Reference recommender
3.2 System implementation
3.2.1 Painting producer
1.Training data
2.Painting generation network
3.2.2 Reference recommender
3.2.3 Edge synthesizer
4 Experiment
4.1 Comparison with the original pix2pixHD method
4.2 System evaluation
4.2.1 Study 1 methodology
4.2.2 Study 1 results
4.2.3 Study 2 methodology
4.2.4 Study 2 results
5 Discussion
6 Conclusions
Compliance with ethics guidelines
本文編號:3791400
【文章頁數(shù)】:14 頁
【文章目錄】:
1 Introduction
2 Related work
2.1 Understanding user inputs
2.2 Turning a sketch into a painting
3 System overview and implementation
3.1 System overview
1.Painting producer
2.Edge synthesizer
3.Reference recommender
3.2 System implementation
3.2.1 Painting producer
1.Training data
2.Painting generation network
3.2.2 Reference recommender
3.2.3 Edge synthesizer
4 Experiment
4.1 Comparison with the original pix2pixHD method
4.2 System evaluation
4.2.1 Study 1 methodology
4.2.2 Study 1 results
4.2.3 Study 2 methodology
4.2.4 Study 2 results
5 Discussion
6 Conclusions
Compliance with ethics guidelines
本文編號:3791400
本文鏈接:http://sikaile.net/wenyilunwen/dongmansheji/3791400.html
最近更新
教材專著