Generation of Stereo Images Based on a View Synthesis Network
The conventional warping method only considers translations of pixels to generate stereo images. In this paper, we propose a model that can generate stereo images from a single image, considering both translation as well as rotation of objects in the image. We modified the appearance flow network to...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-04-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/10/9/3101 |
| id |
doaj-00c239b3cfdb4a24a8a923b112a9ef71 |
|---|---|
| record_format |
Article |
| spelling |
doaj-00c239b3cfdb4a24a8a923b112a9ef712020-11-25T02:11:22ZengMDPI AGApplied Sciences2076-34172020-04-01103101310110.3390/app10093101Generation of Stereo Images Based on a View Synthesis NetworkYuan-Mau Lo0Chin-Chen Chang1Der-Lor Way2Zen-Chung Shih3Institute of Multimedia Engineering, National Chiao Tung University, Hsinchu 300, TaiwanDepartment of Computer Science and Information Engineering, National United University, Miaoli 360, TaiwanDepartment of NewMedia Art, Taipei National University of Arts, Taipei 112, TaiwanInstitute of Multimedia Engineering, National Chiao Tung University, Hsinchu 300, TaiwanThe conventional warping method only considers translations of pixels to generate stereo images. In this paper, we propose a model that can generate stereo images from a single image, considering both translation as well as rotation of objects in the image. We modified the appearance flow network to make it more general and suitable for our model. We also used a reference image to improve the inpainting method. The quality of images resulting from our model is better than that of images generated using conventional warping. Our model also better retained the structure of objects in the input image. In addition, our model does not limit the size of the input image. Most importantly, because our model considers the rotation of objects, the resulting images appear more stereoscopic when viewed with a device.https://www.mdpi.com/2076-3417/10/9/3101stereo imagesview synthesisneural networksemantic segmentationdepth estimation |
| collection |
DOAJ |
| language |
English |
| format |
Article |
| sources |
DOAJ |
| author |
Yuan-Mau Lo Chin-Chen Chang Der-Lor Way Zen-Chung Shih |
| spellingShingle |
Yuan-Mau Lo Chin-Chen Chang Der-Lor Way Zen-Chung Shih Generation of Stereo Images Based on a View Synthesis Network Applied Sciences stereo images view synthesis neural network semantic segmentation depth estimation |
| author_facet |
Yuan-Mau Lo Chin-Chen Chang Der-Lor Way Zen-Chung Shih |
| author_sort |
Yuan-Mau Lo |
| title |
Generation of Stereo Images Based on a View Synthesis Network |
| title_short |
Generation of Stereo Images Based on a View Synthesis Network |
| title_full |
Generation of Stereo Images Based on a View Synthesis Network |
| title_fullStr |
Generation of Stereo Images Based on a View Synthesis Network |
| title_full_unstemmed |
Generation of Stereo Images Based on a View Synthesis Network |
| title_sort |
generation of stereo images based on a view synthesis network |
| publisher |
MDPI AG |
| series |
Applied Sciences |
| issn |
2076-3417 |
| publishDate |
2020-04-01 |
| description |
The conventional warping method only considers translations of pixels to generate stereo images. In this paper, we propose a model that can generate stereo images from a single image, considering both translation as well as rotation of objects in the image. We modified the appearance flow network to make it more general and suitable for our model. We also used a reference image to improve the inpainting method. The quality of images resulting from our model is better than that of images generated using conventional warping. Our model also better retained the structure of objects in the input image. In addition, our model does not limit the size of the input image. Most importantly, because our model considers the rotation of objects, the resulting images appear more stereoscopic when viewed with a device. |
| topic |
stereo images view synthesis neural network semantic segmentation depth estimation |
| url |
https://www.mdpi.com/2076-3417/10/9/3101 |
| work_keys_str_mv |
AT yuanmaulo generationofstereoimagesbasedonaviewsynthesisnetwork AT chinchenchang generationofstereoimagesbasedonaviewsynthesisnetwork AT derlorway generationofstereoimagesbasedonaviewsynthesisnetwork AT zenchungshih generationofstereoimagesbasedonaviewsynthesisnetwork |
| _version_ |
1724914715982823424 |