Search papers, labs, and topics across Lattice.
This paper introduces a quantization-aware image enhancement model tailored for mobile deployment, addressing the performance degradation that occurs when high-quality image enhancement models are quantized for use on mobile devices. The proposed architecture employs a hierarchical network with gated encoder blocks and multiscale refinement to preserve fine-grained features. By incorporating quantization-aware training (QAT), the model learns to mitigate the effects of low-precision representation, resulting in high-fidelity visual output with low computational overhead suitable for mobile devices.
Achieve high-fidelity image enhancement on mobile devices even after quantization by training a model that anticipates and adapts to low-precision representations.
Image enhancement models for mobile devices often struggle to balance high output quality with the fast processing speeds required by mobile hardware. While recent deep learning models can enhance low-quality mobile photos into high-quality images, their performance is often degraded when converted to lower-precision formats for actual use on mobile phones. To address this training-deployment mismatch, we propose an efficient image enhancement model designed specifically for mobile deployment. Our approach uses a hierarchical network architecture with gated encoder blocks and multiscale refinement to preserve fine-grained visual features. Moreover, we incorporate Quantization-Aware Training (QAT) to simulate the effects of low-precision representation during the training process. This allows the network to adapt and prevents the typical drop in quality seen with standard post-training quantization (PTQ). Experimental results demonstrate that the proposed method produces high-fidelity visual output while maintaining the low computational overhead needed for practical use on standard mobile devices. The code will be available at https://github.com/GenAI4E/QATIE.git.
