EUVIMEDEUVIMED entdecken
Crossref

Task-Switchable Pre-Processor for Image Compression for Multiple Machine Vision Tasks

Mingyi Yang, Fei Yang, Luka Murn, Marc Gorriz Blanch, Juil Sock, Shuai Wan, Fuzheng Yang, Luis Herranz

IEEE Transactions on Circuits and Systems for Video Technology · 2024-07 · 34 (7) · 6416-6429

DOI: 10.1109/tcsvt.2023.3348995 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Circuits and Systems for Video Technology, 2024-07. Verfasst von Mingyi Yang und 7 weiteren Autor:innen. Die Arbeit wurde bisher 19-mal zitiert (Crossref) und nennt 63 Literaturangaben. Gefördert von: Ramon y Cajal grant, National Natural Science Foundation of China, Ministerio de Ciencia e Innovaci?n.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Visual content is increasingly being processed by machines for various automated content analysis tasks instead of being consumed by humans. Despite the existence of several compression methods tailored for machine tasks, few consider real-world scenarios with multiple tasks. In this paper, we aim to address this gap by proposing a task-switchable pre-processor that optimizes input images specifically for machine consumption prior to encoding by an off-the-shelf codec designed for human consumption. The proposed task-switchable pre-processor adeptly maintains relevant semantic information based on the specific characteristics of different downstream tasks, while effectively suppressing irrelevant information to reduce bitrate. To enhance the processing of semantic information for diverse tasks, we leverage pre-extracted semantic features to modulate the pixel-to-pixel mapping within the pre-processor. By switching between different modulations, multiple tasks can be seamlessly incorporated into the system. Extensive experiments demonstrate the practicality and simplicity of our approach. It significantly reduces the number of parameters required for handling multiple tasks while still delivering impressive performance. Our method showcases the potential to achieve efficient and effective compression for machine vision tasks, supporting the evolving demands of real-world applications.

Förderung

  • Ramon y Cajal grant
  • National Natural Science Foundation of China
  • Ministerio de Ciencia e Innovaci?n
  • Fundamental Research Funds for the Central Universities

Zitieren

APA

Yang, M., Yang, F., Murn, L., Blanch, M. G., Sock, J., Wan, S., Yang, F., & Herranz, L. (2024). Task-Switchable Pre-Processor for Image Compression for Multiple Machine Vision Tasks. IEEE Transactions on Circuits and Systems for Video Technology, 34(7), 6416-6429. https://doi.org/10.1109/tcsvt.2023.3348995

Vancouver

Yang M, Yang F, Murn L, Blanch MG, Sock J, Wan S, et al. Task-Switchable Pre-Processor for Image Compression for Multiple Machine Vision Tasks. IEEE Transactions on Circuits and Systems for Video Technology. 2024;34(7):6416-6429. doi:10.1109/tcsvt.2023.3348995

BibTeX
@article{Yang2024,
  author = {Yang, Mingyi and Yang, Fei and Murn, Luka and Blanch, Marc Gorriz and Sock, Juil and Wan, Shuai and Yang, Fuzheng and Herranz, Luis},
  title = {Task-Switchable Pre-Processor for Image Compression for Multiple Machine Vision Tasks},
  journal = {IEEE Transactions on Circuits and Systems for Video Technology},
  year = {2024},
  volume = {34},
  number = {7},
  pages = {6416--6429},
  doi = {10.1109/tcsvt.2023.3348995}
}
Kostenlos starten →