Distributed analysis of multilayer signal coding

JP7863549B2Active Publication Date: 2026-05-21V NOVA INT LTD
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
V NOVA INT LTD
Filing Date
2021-10-15
Publication Date
2026-05-21

AI Technical Summary

Technical Problem

Existing methods for AI analysis of high-resolution image or video feeds from remote locations face challenges in efficient data transfer and processing, as they often require high-bandwidth connectivity or local feature extraction, which is inflexible and costly, and lack guidance on useful features for analysis.

Method used

Implementing multilayer coding schemes like VC-6 and LCEVC to encode signals at the edge, transmitting lower-quality layers for initial analysis, and requesting higher-quality layers on demand for enhanced processing, enabling efficient data transmission and flexible AI processing.

Benefits of technology

This approach allows for efficient transmission of high-frequency signal streams over constrained bandwidth, supports nearly consecutive frame analysis, and enhances classification accuracy by selectively requesting higher-resolution data, improving AI processing efficiency and flexibility.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 0007863549000001
    Figure 0007863549000001
  • Figure 0007863549000002
    Figure 0007863549000002
  • Figure 0007863549000003
    Figure 0007863549000003
Patent Text Reader

Abstract

The present disclosure provides a method for analyzing multiple video camera feeds, the method comprising: encoding, at a first location, the multiple video camera feeds using layer-based coding, including generating an encoded data stream for each of multiple layers in the layer-based coding, where different layers in the multiple layers correspond to different spatial resolutions, with higher layers representing higher spatial resolutions; transmitting, from the first location to a remote second location, the encoded data streams for one or more lowest layers of the multiple video camera feeds; and, at the second location, decoding the encoded data streams to generate a set of reconstructions of the multiple video camera feeds at the first spatial resolution. applying one or more video analytics functions to the identified one or more video camera feeds to identify one or more video camera feeds for further analysis; transmitting a request for additional encoded data streams of one or more layers above the one or more lowest layers to a first location of the identified one or more video camera feeds for further analysis; transmitting the additional encoded data streams of one or more layers above the one or more lowest layers to a second location in response to the request; decoding the additional encoded data streams at the second location to generate a set of reconstructions of the identified one or more video camera feeds at a second spatial resolution; and applying the one or more video analytics functions to the set of reconstructions at the second spatial resolution.
Need to check novelty before this filing date? Find Prior Art