Taming Lookup Tables for Efficient Image Retouching
Preprint
Abstract
The widespread use of high-definition screens in edge devices, such as end-user cameras, smartphones, and televisions, is spurring a significant demand for image enhancement. Existing enhancement models often optimize for high performance while falling short of reducing hardware inference time and power consumption, especially on edge devices with constrained computing and storage resources. To this end, we propose Image Color Enhancement Lookup Table (ICELUT) that adopts LUTs for extremely efficient edge inference, without any convolutional neural network (CNN). During training, we leverage pointwise (1x1) convolution to extract color information, alongside a split fully connected layer to incorporate global information. Both components are then seamlessly converted into LUTs for hardware-agnostic deployment. ICELUT achieves near-state-of-the-art performance and remarkably low power consumption. We observe that the pointwise network structure exhibits robust scalability, upkeeping the performance even with a heavily downsampled 32x32 input image. These enable ICELUT, the first-ever purely LUT-based image enhancer, to reach an unprecedented speed of 0.4ms on GPU and 7ms on CPU, at least one order faster than any CNN solution. Codes are available at https://github.com/Stephen0808/ICELUT.Keywords
All Related Versions
This publication has 6 references indexed in Scilit:
- Hundred-Kilobyte Lookup Tables for Efficient Single-Image Super-Resolution2024
- 4D LUT: Learnable Context-Aware 4D Lookup Table for Image EnhancementIEEE Transactions on Image Processing, 2023
- Learning Image-adaptive 3D Lookup Tables for High Performance Photo Enhancement in Real-timeIEEE Transactions on Pattern Analysis and Machine Intelligence, 2020
- Deep bilateral learning for real-time image enhancementACM Transactions on Graphics, 2017
- Data-driven hallucination of different times of day from a single outdoor photoACM Transactions on Graphics, 2013
- Image quality assessment: from error visibility to structural similarityIEEE Transactions on Image Processing, 2004