Information Fusion based Quality Enhancement for 3D Stereo Images Using CNN. Jin, Z., Luo, H., Luo, L., Zou, W., Li, X., & Steinbach, E. In 2018 26th European Signal Processing Conference (EUSIPCO), pages 1447-1451, Sep., 2018.
![Information Fusion based Quality Enhancement for 3D Stereo Images Using CNN [pdf]](https://bibbase.org/img/filetypes/pdf.svg "pdf")
Paper doi abstract bibtex
Paper doi abstract bibtex Stereo images provide users with a vivid 3D watching experience. Supported by per-view depth maps, 3D stereo images can be used to generate any required intermediate view between the given left and right stereo views. However, 3D stereo images lead to higher transmission and storage cost compared to single view images. Based on the binocular suppression theory, mixed-quality stereo images can alleviate this problem by employing different compression ratios on the two views. This causes noticeable visual artifacts when a high compression ratio is adopted and limits free-viewpoint applications. Hence, the low quality image at the receiver side needs to be enhanced to match the high quality one. To address this problem, in this paper we propose an end-to-end fully Convolutional Neural Network (CNN) for enhancing the low quality images in quality asymmetric stereo images by exploiting inter-view correlation. The proposed network achieves an image quality boost of up to 4.6dB and 3.88dB PSNR gain over ordinary JPEG for QF10 and 20, respectively, and an improvement of up to 2.37dB and 2.05dB over the state-of-the-art CNN-based results for QF10 and 20, respectively.
@InProceedings{8553082,
author = {Z. Jin and H. Luo and L. Luo and W. Zou and X. Li and E. Steinbach},
booktitle = {2018 26th European Signal Processing Conference (EUSIPCO)},
title = {Information Fusion based Quality Enhancement for 3D Stereo Images Using CNN},
year = {2018},
pages = {1447-1451},
abstract = {Stereo images provide users with a vivid 3D watching experience. Supported by per-view depth maps, 3D stereo images can be used to generate any required intermediate view between the given left and right stereo views. However, 3D stereo images lead to higher transmission and storage cost compared to single view images. Based on the binocular suppression theory, mixed-quality stereo images can alleviate this problem by employing different compression ratios on the two views. This causes noticeable visual artifacts when a high compression ratio is adopted and limits free-viewpoint applications. Hence, the low quality image at the receiver side needs to be enhanced to match the high quality one. To address this problem, in this paper we propose an end-to-end fully Convolutional Neural Network (CNN) for enhancing the low quality images in quality asymmetric stereo images by exploiting inter-view correlation. The proposed network achieves an image quality boost of up to 4.6dB and 3.88dB PSNR gain over ordinary JPEG for QF10 and 20, respectively, and an improvement of up to 2.37dB and 2.05dB over the state-of-the-art CNN-based results for QF10 and 20, respectively.},
keywords = {data compression;feedforward neural nets;image coding;image reconstruction;stereo image processing;required intermediate view;stereo views;single view images;mixed-quality stereo images;low quality image;quality asymmetric stereo images;image quality boost;quality enhancement;vivid 3D watching experience;per-view depth maps;PSNR gain;Image coding;Transform coding;Convolution;Feature extraction;Image reconstruction;Hafnium;Three-dimensional displays},
doi = {10.23919/EUSIPCO.2018.8553082},
issn = {2076-1465},
month = {Sep.},
url = {https://www.eurasip.org/proceedings/eusipco/eusipco2018/papers/1570437356.pdf},
}Downloads: 0
{"_id":"uSrqjW85egFLhHNHo","bibbaseid":"jin-luo-luo-zou-li-steinbach-informationfusionbasedqualityenhancementfor3dstereoimagesusingcnn-2018","authorIDs":[],"author_short":["Jin, Z.","Luo, H.","Luo, L.","Zou, W.","Li, X.","Steinbach, E."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","author":[{"firstnames":["Z."],"propositions":[],"lastnames":["Jin"],"suffixes":[]},{"firstnames":["H."],"propositions":[],"lastnames":["Luo"],"suffixes":[]},{"firstnames":["L."],"propositions":[],"lastnames":["Luo"],"suffixes":[]},{"firstnames":["W."],"propositions":[],"lastnames":["Zou"],"suffixes":[]},{"firstnames":["X."],"propositions":[],"lastnames":["Li"],"suffixes":[]},{"firstnames":["E."],"propositions":[],"lastnames":["Steinbach"],"suffixes":[]}],"booktitle":"2018 26th European Signal Processing Conference (EUSIPCO)","title":"Information Fusion based Quality Enhancement for 3D Stereo Images Using CNN","year":"2018","pages":"1447-1451","abstract":"Stereo images provide users with a vivid 3D watching experience. Supported by per-view depth maps, 3D stereo images can be used to generate any required intermediate view between the given left and right stereo views. However, 3D stereo images lead to higher transmission and storage cost compared to single view images. Based on the binocular suppression theory, mixed-quality stereo images can alleviate this problem by employing different compression ratios on the two views. This causes noticeable visual artifacts when a high compression ratio is adopted and limits free-viewpoint applications. Hence, the low quality image at the receiver side needs to be enhanced to match the high quality one. To address this problem, in this paper we propose an end-to-end fully Convolutional Neural Network (CNN) for enhancing the low quality images in quality asymmetric stereo images by exploiting inter-view correlation. The proposed network achieves an image quality boost of up to 4.6dB and 3.88dB PSNR gain over ordinary JPEG for QF10 and 20, respectively, and an improvement of up to 2.37dB and 2.05dB over the state-of-the-art CNN-based results for QF10 and 20, respectively.","keywords":"data compression;feedforward neural nets;image coding;image reconstruction;stereo image processing;required intermediate view;stereo views;single view images;mixed-quality stereo images;low quality image;quality asymmetric stereo images;image quality boost;quality enhancement;vivid 3D watching experience;per-view depth maps;PSNR gain;Image coding;Transform coding;Convolution;Feature extraction;Image reconstruction;Hafnium;Three-dimensional displays","doi":"10.23919/EUSIPCO.2018.8553082","issn":"2076-1465","month":"Sep.","url":"https://www.eurasip.org/proceedings/eusipco/eusipco2018/papers/1570437356.pdf","bibtex":"@InProceedings{8553082,\n author = {Z. Jin and H. Luo and L. Luo and W. Zou and X. Li and E. Steinbach},\n booktitle = {2018 26th European Signal Processing Conference (EUSIPCO)},\n title = {Information Fusion based Quality Enhancement for 3D Stereo Images Using CNN},\n year = {2018},\n pages = {1447-1451},\n abstract = {Stereo images provide users with a vivid 3D watching experience. Supported by per-view depth maps, 3D stereo images can be used to generate any required intermediate view between the given left and right stereo views. However, 3D stereo images lead to higher transmission and storage cost compared to single view images. Based on the binocular suppression theory, mixed-quality stereo images can alleviate this problem by employing different compression ratios on the two views. This causes noticeable visual artifacts when a high compression ratio is adopted and limits free-viewpoint applications. Hence, the low quality image at the receiver side needs to be enhanced to match the high quality one. To address this problem, in this paper we propose an end-to-end fully Convolutional Neural Network (CNN) for enhancing the low quality images in quality asymmetric stereo images by exploiting inter-view correlation. The proposed network achieves an image quality boost of up to 4.6dB and 3.88dB PSNR gain over ordinary JPEG for QF10 and 20, respectively, and an improvement of up to 2.37dB and 2.05dB over the state-of-the-art CNN-based results for QF10 and 20, respectively.},\n keywords = {data compression;feedforward neural nets;image coding;image reconstruction;stereo image processing;required intermediate view;stereo views;single view images;mixed-quality stereo images;low quality image;quality asymmetric stereo images;image quality boost;quality enhancement;vivid 3D watching experience;per-view depth maps;PSNR gain;Image coding;Transform coding;Convolution;Feature extraction;Image reconstruction;Hafnium;Three-dimensional displays},\n doi = {10.23919/EUSIPCO.2018.8553082},\n issn = {2076-1465},\n month = {Sep.},\n url = {https://www.eurasip.org/proceedings/eusipco/eusipco2018/papers/1570437356.pdf},\n}\n\n","author_short":["Jin, Z.","Luo, H.","Luo, L.","Zou, W.","Li, X.","Steinbach, E."],"key":"8553082","id":"8553082","bibbaseid":"jin-luo-luo-zou-li-steinbach-informationfusionbasedqualityenhancementfor3dstereoimagesusingcnn-2018","role":"author","urls":{"Paper":"https://www.eurasip.org/proceedings/eusipco/eusipco2018/papers/1570437356.pdf"},"keyword":["data compression;feedforward neural nets;image coding;image reconstruction;stereo image processing;required intermediate view;stereo views;single view images;mixed-quality stereo images;low quality image;quality asymmetric stereo images;image quality boost;quality enhancement;vivid 3D watching experience;per-view depth maps;PSNR gain;Image coding;Transform coding;Convolution;Feature extraction;Image reconstruction;Hafnium;Three-dimensional displays"],"metadata":{"authorlinks":{}},"downloads":0},"bibtype":"inproceedings","biburl":"https://raw.githubusercontent.com/Roznn/EUSIPCO/main/eusipco2018url.bib","creationDate":"2021-02-13T15:38:40.108Z","downloads":0,"keywords":["data compression;feedforward neural nets;image coding;image reconstruction;stereo image processing;required intermediate view;stereo views;single view images;mixed-quality stereo images;low quality image;quality asymmetric stereo images;image quality boost;quality enhancement;vivid 3d watching experience;per-view depth maps;psnr gain;image coding;transform coding;convolution;feature extraction;image reconstruction;hafnium;three-dimensional displays"],"search_terms":["information","fusion","based","quality","enhancement","stereo","images","using","cnn","jin","luo","luo","zou","li","steinbach"],"title":"Information Fusion based Quality Enhancement for 3D Stereo Images Using CNN","year":2018,"dataSources":["yiZioZximP7hphDpY","iuBeKSmaES2fHcEE9"]}