Easz: An Agile Transformer-based Image Compression Framework for Resource-constrained IoTs

2025-05-03

AI Summary

To address the deployment challenges of neural image compression on resource-constrained edge devices, this paper proposes a cloud-edge collaborative lightweight image compression framework: the cloud server performs patch erasure via conditional uniform sampling, while the edge device executes only an ultra-lightweight Transformer for reconstruction—achieving a “zero-edge-computation” architecture. The method supports dynamic bitrate switching, with decoder parameters under 50K and minimal inference latency. Evaluation on a real-world IoT testbed demonstrates zero computational overhead at the encoder, 67% reduction in decoding latency, 2.1 dB PSNR gain, and millisecond-level bitrate adaptation. Key contributions include: (i) the first zero-computation paradigm for edge devices; (ii) a controllable patch erasure mechanism driven by conditional uniform sampling; and (iii) a dynamically adaptive reconstruction design optimized for ultra-low-power scenarios.

Technology Category

• Machine Learning: Learning on the Edge & Model Compression
• Computer Vision: Learning & Optimization for CV
• Data Mining & Knowledge Management: Data Compression

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Cloud, edge and content delivery systems for the Web
• Economics, Online Markets and Human Computation: Architectures and workflows that use LLMs for crowd work
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

Neural image compression, necessary in various machine-to-machine communication scenarios, suffers from its heavy encode-decode structures and inflexibility in switching between different compression levels. Consequently, it raises significant challenges in applying the neural image compression to edge devices that are developed for powerful servers with high computational and storage capacities. We take a step to solve the challenges by proposing a new transformer-based edge-compute-free image coding framework called Easz. Easz shifts the computational overhead to the server, and hence avoids the heavy encoding and model switching overhead on the edge. Easz utilizes a patch-erase algorithm to selectively remove image contents using a conditional uniform-based sampler. The erased pixels are reconstructed on the receiver side through a transformer-based framework. To further reduce the computational overhead on the receiver, we then introduce a lightweight transformer-based reconstruction structure to reduce the reconstruction load on the receiver side. Extensive evaluations conducted on a real-world testbed demonstrate multiple advantages of Easz over existing compression approaches, in terms of adaptability to different compression levels, computational efficiency, and image reconstruction quality.

Problem

• Neural image compression struggles with heavy encode-decode structures.
• Edge devices lack flexibility for varying compression levels.
• Existing methods demand high computational and storage resources.

Innovation

• Transformer-based edge-compute-free image coding framework
• Patch-erase algorithm for selective content removal
• Lightweight transformer-based reconstruction structure