DynamicISP: Dynamically Controlled Image Signal Processor for Image Recognition
This addresses the trade-off between expressive power and computational cost in ISP tuning for edge devices, offering an incremental improvement over existing methods.
The paper tackles the problem of sub-optimal manual tuning of Image Signal Processor (ISP) parameters for image recognition by proposing DynamicISP, which dynamically controls ISP parameters based on previous frame recognition results, achieving state-of-the-art accuracy with low computational cost in object detection tasks.
Image Signal Processors (ISPs) play important roles in image recognition tasks as well as in the perceptual quality of captured images. In most cases, experts make a lot of effort to manually tune many parameters of ISPs, but the parameters are sub-optimal. In the literature, two types of techniques have been actively studied: a machine learning-based parameter tuning technique and a DNN-based ISP technique. The former is lightweight but lacks expressive power. The latter has expressive power, but the computational cost is too heavy on edge devices. To solve these problems, we propose "DynamicISP," which consists of multiple classical ISP functions and dynamically controls the parameters of each frame according to the recognition result of the previous frame. We show our method successfully controls the parameters of multiple ISP functions and achieves state-of-the-art accuracy with low computational cost in single and multi-category object detection tasks.