Lookup Register-Tables with Interpolation for Effective Image Transformation on x86/64 CPUs

Lookup tables (LUTs) are commonly used to speed up image processing by handling complex mathematical functions like sine and exponential calculations. They are used in various applications such as camera image processing, high-dynamic range imaging, and edge-preserving filtering. However, due to the increasing gap between computing and input/output performance, LUTs are becoming less effective. Even though specific circuits like SIMD can improve LUT efficiency, they still need to bridge the performance gap fully. The gap makes it difficult to choose between direct numerical and LUT calculations. For this problem, a register-LUTs method with the nearest neighbor was proposed; however, it is limited for functions with narrow-range values approaching zero. In this paper, we propose a method for using register LUTs to process images efficiently over a wide range of values. Our contributions include proposing register-LUT with linear interpolation for efficient computation, using a smaller data type for further efficiency, and suggesting an efficient data retrieving method.

I. Introduction

Lookup tables (LUTs) are a fundamental acceleration method for image processing. The LUT acceleration is usually used in the image-to-image functional transformation, which uses costly transcendental functions (e.g., sin, cos, exp, log, and pow). For example, contrast transformation in-camera image signal processing (ISP) [1], high-dynamic range imaging [2], activation functions in neural networks [3], and acceleration of edge-preserving filtering, such as bilateral filtering [4] and local Laplacian filtering [5]. For the acceleration, images are decomposed by transformation functions, which are interpolation [2], [6]–[8], polynomial decompisiton [9], [10], raised cosine decompision [11]–[14], Fourier transform [15]–[18], and eigen/singular value decomposition [19], [20]. Also pixelwise-LUT, i.e., 3D LUT is used for various image processing as learning-based [21]–[25] and non-lerning one [26].