Edge-Cloud Collaborative Semantic Image Compression and Generative Reconstruction for Low-Bandwidth Transmission

Authors:
Open Access
Publisher website
Google Scholar
Abstract
In bandwidth-limited image transmission, maintaining perceptual quality at very low bit rates remains challenging. Conventional compression codecs typically introduce noticeable blocking artifacts and discard fine details when the compression ratio becomes too aggressive. To address this problem, we present BCP-SDXL, an edge-cloud collaborative image compression framework with cloud-side generative reconstruction. Instead of transmitting full-resolution pixels, BCP-SDXL sends two lightweight cues: a compressed Base image and a Canny structural map. On the edge side, an Adaptive Compression Agent, built on a Deep Q-Network (DQN), dynamically selects compression parameters and extracts structural priors with low deployment overhead. Upon reception, the cloud server leverages the Multimodal Large Language Model (Qwen3-VL-2B) to analyze the compressed base image and generate high-fidelity semantic descriptions. These descriptions, together with the base image and edge maps, are then used to condition a Latent Diffusion Model (SDXL) integrated with ControlNet. The resulting multi-condition reconstruction process restores missing high-frequency details while preserving the original scene’s structural layout. Experiments on public datasets show that BCP-SDXL reduces transmission cost by more than 90% and achieves better perceptual quality than conventional codecs and recent deep learning baselines under highly compressed settings.
Funding Information
  • National Key Research and Development Plan (2019YFB2101900)
  • Tianjin Natural Science Foundation Project (25JCZDJC00250)
  • Natural Science Foundation of China (61602345)
  • Joint Fund of the National Natural Science Foundation of China and the Karst Science Research Center of Guizhou Province (62002263)
  • TianKai Higher Education Innovation Park Enterprise R&D Special Project (23YFZXYC00046)

This publication has 12 references indexed in Scilit: