Video coding for machines using region-of-interest-based retargeting
Abstract The work is focused on video compression for the scenarios, where the decoded video serves not only human viewers but also as input for systems implementing various machine vision tasks, such as object detection and tracking. The proposed innovative tool is based on retargeting video frames processed based on Regions of Interest (RoI), corresponding to the individual objects detected in the frames. Experimental evaluation demonstrates significant average bitrate reduction while maintaining the same quality, ranging from 3 to 57% depending on the machine vision task and encoding scenario. The proposal underwent thorough consideration within the MPEG group and was adopted for the upcoming Video Coding for Machines (VCM) technology.
Authors
- Jakub Stankowski (ORCID: https://orcid.org/0000-0002-5105-5090)
- Mateusz Lorkiewicz (ORCID: https://orcid.org/0000-0002-9020-1927)
- M. Wawrzyniak (ORCID: https://orcid.org/0000-0002-2877-4142)
- Sławomir Maćkowiak (ORCID: https://orcid.org/0000-0003-4534-3827)
- Tomasz Grajek (ORCID: https://orcid.org/0000-0002-7142-9416)
- Marek Domański (ORCID: https://orcid.org/0000-0002-9381-0293)
- Olgierd Stankiewicz (ORCID: https://orcid.org/0000-0001-9691-9094)
- Sławomir Różek
- Ding Ding (ORCID: https://orcid.org/0009-0009-3873-1195)
- Shan Liu (ORCID: https://orcid.org/0000-0003-1847-8769)
Institutions
- Poznań University of Technology (PL)
- KLA (United States) (US)
Publication Details
- Journal
- EURASIP Journal on Image and Video Processing
- Published
- 2025-10-24
- DOI
- https://doi.org/10.1186/s13640-025-00682-3
- Primary Topic
- Advanced Image and Video Retrieval Techniques
- Type
- article
- Field-Weighted Citation Impact
- 0.00