Edge deployment-oriented neural network video coding model lightweight method

By introducing structured sparsity constraints and feature guidance mechanisms into the neural network video coding model, redundant structures are automatically identified and reduced, solving the problem of high computing resources and storage overhead on edge devices, and achieving efficient video compression and preservation of reconstruction quality.

CN121967701APending Publication Date: 2026-05-01ZHUHAI XINLINGYU COMMUNICATIONS CO LTD
0 Cites 0 Cited by

Patent Information

Application Number
CN202610164091.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-02-05
Publication Date
2026-05-01

AI Technical Summary

Technical Problem

Existing neural network video coding models incur high computational and storage overhead when deployed on edge devices, and existing lightweight technologies struggle to maintain stability and efficiency in terms of reconstruction quality and channel correlation.

Method used

By introducing structured sparsity constraints and feature guidance mechanisms into the model, and using sparsity loss function and rate distortion loss to jointly train the mask parameters, redundant structures are automatically identified and reduced. Combined with the features of the target benchmark model for collaborative training, channel-level structure optimization is achieved.

Benefits of technology

It significantly reduces computing and storage costs on edge devices while maintaining good compression efficiency and video quality, improving the feasibility of deploying the model in resource-constrained environments.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention discloses an edge deployment-oriented neural network video coding model lightweight method, which comprises the following steps of: constructing a video compression model as a reference model to be optimized; inserting a dry mask layer which can be trained into a preset position of a reference model to be optimized, and performing joint training on mask parameters by utilizing a sparse loss function and rate distortion loss to obtain mask parameter distribution; when the proportion of the mask parameter compressed to be zero in any mask layer reaches a preset target sparsity, sparse updating of the corresponding mask layer is stopped, and a mask layer state reaching a sparsity target is obtained; performing fusion operation on each mask layer and the adjacent network layer to obtain a neural network model after structure compression; and removing all mask layers to obtain a lightweight video coding model, and deploying the lightweight video coding model on edge equipment to execute a video compression task. According to the method, the model complexity can be greatly reduced, and the reasoning speed and the deployment efficiency of the edge equipment are improved.
Need to check novelty before this filing date? Find Prior Art