Distortion-Resilient DIBR for Novel View Synthesis from a Single Image

Yuchen Liu; Eiji Kamioka; Phan Xuan Tan

doi:10.1007/978-981-96-4282-3_24

Distortion-Resilient DIBR for Novel View Synthesis from a Single Image

研究成果: Conference contribution

抄録

It is challenging to render novel views from a single image input due to inherent ambiguities in the geometry and texture information of the desired scene. As a consequence, existing methods often encounter various types of distortions in synthesized views. To this end, we propose a distortion-resilient Depth-Image-Based Rendering (DIBR) method for synthesizing novel views given a single image input. The proposed method is qualitatively and quantitatively evaluated on the Real-Estate 10K dataset, showing superior results compared to baselines.

本文言語	English
ホスト出版物のタイトル	Information and Communication Technology - 13th International Symposium, SOICT 2024, Proceedings
編集者	Wray Buntine, Morten Fjeld, Truyen Tran, Minh-Triet Tran, Binh Huynh Thi Thanh, Takumi Miyoshi
出版社	Springer Science and Business Media Deutschland GmbH
ページ	287-297
ページ数	11
ISBN（印刷版）	9789819642816
DOI	https://doi.org/10.1007/978-981-96-4282-3_24
出版ステータス	Published - 2025
イベント	13th International Symposium on Information and Communication Technology, SOICT 2024 - Danang, Viet Nam 継続期間: 2024 12月 13 → 2024 12月 15

出版物シリーズ

名前	Communications in Computer and Information Science
巻	2350 CCIS
ISSN（印刷版）	1865-0929
ISSN（電子版）	1865-0937

Conference

Conference	13th International Symposium on Information and Communication Technology, SOICT 2024
国/地域	Viet Nam
City	Danang
Period	24/12/13 → 24/12/15

ASJC Scopus subject areas

コンピュータサイエンス一般
数学一般

文献へのアクセス

10.1007/978-981-96-4282-3_24

他のファイルとリンク

引用スタイル

Liu, Y., Kamioka, E., & Tan, P. X. (2025). Distortion-Resilient DIBR for Novel View Synthesis from a Single Image. In W. Buntine, M. Fjeld, T. Tran, M.-T. Tran, B. Huynh Thi Thanh, & T. Miyoshi (Eds.), Information and Communication Technology - 13th International Symposium, SOICT 2024, Proceedings (pp. 287-297). (Communications in Computer and Information Science; Vol. 2350 CCIS). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-981-96-4282-3_24

Distortion-Resilient DIBR for Novel View Synthesis from a Single Image. / Liu, Yuchen; Kamioka, Eiji ; Tan, Phan Xuan.
Information and Communication Technology - 13th International Symposium, SOICT 2024, Proceedings. ed. / Wray Buntine; Morten Fjeld; Truyen Tran; Minh-Triet Tran; Binh Huynh Thi Thanh; Takumi Miyoshi. Springer Science and Business Media Deutschland GmbH, 2025. p. 287-297 (Communications in Computer and Information Science; Vol. 2350 CCIS).

研究成果: Conference contribution

Liu, Y, Kamioka, E & Tan, PX 2025, Distortion-Resilient DIBR for Novel View Synthesis from a Single Image. in W Buntine, M Fjeld, T Tran, M-T Tran, B Huynh Thi Thanh & T Miyoshi (eds), Information and Communication Technology - 13th International Symposium, SOICT 2024, Proceedings. Communications in Computer and Information Science, vol. 2350 CCIS, Springer Science and Business Media Deutschland GmbH, pp. 287-297, 13th International Symposium on Information and Communication Technology, SOICT 2024, Danang, Viet Nam, 24/12/13. https://doi.org/10.1007/978-981-96-4282-3_24

Liu Y, Kamioka E , Tan PX. Distortion-Resilient DIBR for Novel View Synthesis from a Single Image. In Buntine W, Fjeld M, Tran T, Tran MT, Huynh Thi Thanh B, Miyoshi T, editors, Information and Communication Technology - 13th International Symposium, SOICT 2024, Proceedings. Springer Science and Business Media Deutschland GmbH. 2025. p. 287-297. (Communications in Computer and Information Science). doi: 10.1007/978-981-96-4282-3_24

Liu, Yuchen ; Kamioka, Eiji ; Tan, Phan Xuan. / Distortion-Resilient DIBR for Novel View Synthesis from a Single Image. Information and Communication Technology - 13th International Symposium, SOICT 2024, Proceedings. editor / Wray Buntine ; Morten Fjeld ; Truyen Tran ; Minh-Triet Tran ; Binh Huynh Thi Thanh ; Takumi Miyoshi. Springer Science and Business Media Deutschland GmbH, 2025. pp. 287-297 (Communications in Computer and Information Science).

@inproceedings{c7714711ac0c4f38b8511f0d88ccb56a,

title = "Distortion-Resilient DIBR for Novel View Synthesis from a Single Image",

abstract = "It is challenging to render novel views from a single image input due to inherent ambiguities in the geometry and texture information of the desired scene. As a consequence, existing methods often encounter various types of distortions in synthesized views. To this end, we propose a distortion-resilient Depth-Image-Based Rendering (DIBR) method for synthesizing novel views given a single image input. The proposed method is qualitatively and quantitatively evaluated on the Real-Estate 10K dataset, showing superior results compared to baselines.",

keywords = "DIBR, Distortion Handling, Novel View Synthesis",

author = "Yuchen Liu and Eiji Kamioka and Tan, \{Phan Xuan\}",

note = "Publisher Copyright: {\textcopyright} The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.; 13th International Symposium on Information and Communication Technology, SOICT 2024 ; Conference date: 13-12-2024 Through 15-12-2024",

year = "2025",

doi = "10.1007/978-981-96-4282-3\_24",

language = "English",

isbn = "9789819642816",

series = "Communications in Computer and Information Science",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "287--297",

editor = "Wray Buntine and Morten Fjeld and Truyen Tran and Minh-Triet Tran and \{Huynh Thi Thanh\}, Binh and Takumi Miyoshi",

booktitle = "Information and Communication Technology - 13th International Symposium, SOICT 2024, Proceedings",

}

TY - GEN

T1 - Distortion-Resilient DIBR for Novel View Synthesis from a Single Image

AU - Liu, Yuchen

AU - Kamioka, Eiji

AU - Tan, Phan Xuan

N1 - Publisher Copyright: © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

PY - 2025

Y1 - 2025

N2 - It is challenging to render novel views from a single image input due to inherent ambiguities in the geometry and texture information of the desired scene. As a consequence, existing methods often encounter various types of distortions in synthesized views. To this end, we propose a distortion-resilient Depth-Image-Based Rendering (DIBR) method for synthesizing novel views given a single image input. The proposed method is qualitatively and quantitatively evaluated on the Real-Estate 10K dataset, showing superior results compared to baselines.

AB - It is challenging to render novel views from a single image input due to inherent ambiguities in the geometry and texture information of the desired scene. As a consequence, existing methods often encounter various types of distortions in synthesized views. To this end, we propose a distortion-resilient Depth-Image-Based Rendering (DIBR) method for synthesizing novel views given a single image input. The proposed method is qualitatively and quantitatively evaluated on the Real-Estate 10K dataset, showing superior results compared to baselines.

KW - DIBR

KW - Distortion Handling

KW - Novel View Synthesis

UR - https://www.scopus.com/pages/publications/105003858082

UR - https://www.scopus.com/pages/publications/105003858082#tab=citedBy

U2 - 10.1007/978-981-96-4282-3_24

DO - 10.1007/978-981-96-4282-3_24

M3 - Conference contribution

AN - SCOPUS:105003858082

SN - 9789819642816

T3 - Communications in Computer and Information Science

SP - 287

EP - 297

BT - Information and Communication Technology - 13th International Symposium, SOICT 2024, Proceedings

A2 - Buntine, Wray

A2 - Fjeld, Morten

A2 - Tran, Truyen

A2 - Tran, Minh-Triet

A2 - Huynh Thi Thanh, Binh

A2 - Miyoshi, Takumi

PB - Springer Science and Business Media Deutschland GmbH

T2 - 13th International Symposium on Information and Communication Technology, SOICT 2024

Y2 - 13 December 2024 through 15 December 2024

ER -