Projects per year
Abstract
Template-generated videos (TGVs), created by applying animation templates to static images, have become increasingly prevalent, producing massive user-generated content with highly consistent motion patterns. However, existing video compression schemes are designed to eliminate motion redundancy within individual videos, while overlooking the shared motion patterns widespread across TGVs. To address this limitation, we propose a novel compression scheme that effectively leverages inter-video motion priors to enhance the compression efficiency of TGVs. Specifically, the proposed scheme operates as a two-stage pipeline. In the first stage, high-quality motion priors are identified from a representative TGV based on spatial texture and prediction error. In the second stage, these motion priors are intelligently integrated to expand the motion representation space beyond the local candidate lists in Merge and AMVP modes, thereby enabling the codec to remove inter-video redundancy. Experimental results on the versatile video coding test model (VTM-23.0) demonstrate consistent coding gains across various compression scenarios for TGVs, achieving average BD-rate savings of $1.07 1.38 %$, and 1.18% under low-delay P (LDP), low-delay B (LDB), and random access (RA) configurations, respectively.
| Original language | English |
|---|---|
| Title of host publication | 2026 Data Compression Conference, DCC 2026: Proceedings |
| Editors | Ali BILGIN, James E. FOWLER, Joan SERRA-SAGRISTA, Yan YE, James A. STORER |
| Publisher | IEEE Computer Society |
| Pages | 203-212 |
| Number of pages | 10 |
| ISBN (Electronic) | 9798331582616 |
| DOIs | |
| Publication status | Published - 1 Mar 2026 |
| Event | 2026 Data Compression Conference (DCC) - Snowbird, United States Duration: 24 Mar 2026 → 27 Mar 2026 |
Publication series
| Name | Data Compression Conference Proceedings |
|---|---|
| Publisher | IEEE |
| ISSN (Print) | 1068-0314 |
| ISSN (Electronic) | 2375-0359 |
Conference
| Conference | 2026 Data Compression Conference (DCC) |
|---|---|
| Abbreviated title | DCC 2026 |
| Country/Territory | United States |
| City | Snowbird |
| Period | 24/03/26 → 27/03/26 |
Bibliographical note
Publisher Copyright:© 2026 IEEE.
Funding
This work was supported in part by the National Key R&D Program of China (2023YFA1008500), the National Natural Science Foundation of China (NSFC) under grants 62502116 and U22B2035, China Post-Doctoral Science Foundation under Grant 2025M774315, and the Lingnan University Start-up Grant SUG-007/2425 and Faculty Research Grant SDS24A6, SDS24A18.
Fingerprint
Dive into the research topics of 'An Effective Template-Generated Video Compression Scheme by Exploiting Inter-Video Motion Correlation'. Together they form a unique fingerprint.-
Generative Artificial Intelligence-based Visual Data Compression
WANG, M. (PI)
1/04/25 → 31/03/26
Project: Grant Research
-
Intelligent and Efficient Transcoding for the Versatile Video Coding
WANG, M. (PI)
1/11/25 → 15/08/26
Project: Grant Research
-
High efficiency compression for visual scene data
WANG, M. (PI)
16/02/25 → 31/07/26
Project: Grant Research
Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver