FrameCorr: Adaptive, Autoencoder-based Neural Compression for Video Reconstruction in Resource and Timing Constrained Network Settings

John Li, Deepak Nair, Klara Nahrstedt, Indranil Gupta, Shehab Sarar Ahmed

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Video processing is becoming increasingly popular and cost-effective on IoT devices but faces challenges in transmitting data under varying timing constraints and network bandwidth. Existing compression methods struggle with incomplete data. We present FrameCorr, a deep learning framework that leverages prior-received video data to predict and reconstruct missing frame segments, enabling video reconstruction despite data loss.

Original languageEnglish (US)
Title of host publicationProceedings - 2024 International Symposium on Multimedia, ISM 2024
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages197-200
Number of pages4
ISBN (Electronic)9798331511111
DOIs
StatePublished - 2024
Event26th International Symposium on Multimedia, ISM 2024 - Tokyo, Japan
Duration: Dec 11 2024Dec 13 2024

Publication series

NameProceedings - 2024 International Symposium on Multimedia, ISM 2024

Conference

Conference26th International Symposium on Multimedia, ISM 2024
Country/TerritoryJapan
CityTokyo
Period12/11/2412/13/24

Keywords

  • Autoencoder
  • IoT
  • Neural Compression
  • Video Transmission

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Networks and Communications
  • Computer Vision and Pattern Recognition
  • Media Technology

Fingerprint

Dive into the research topics of 'FrameCorr: Adaptive, Autoencoder-based Neural Compression for Video Reconstruction in Resource and Timing Constrained Network Settings'. Together they form a unique fingerprint.

Cite this