Keypoint Detection and Description for Raw Bayer Images
This work addresses efficiency in robotic vision systems for resource-constrained environments, representing a novel but domain-specific incremental improvement.
The paper tackles keypoint detection and description for robotic perception by proposing a network that directly processes raw Bayer images, bypassing the ISP to reduce hardware and memory needs, and it outperforms existing methods on raw images with higher accuracy and stability under rotations and scale variations.
Keypoint detection and local feature description are fundamental tasks in robotic perception, critical for applications such as SLAM, robot localization, feature matching, pose estimation, and 3D mapping. While existing methods predominantly operate on RGB images, we propose a novel network that directly processes raw images, bypassing the need for the Image Signal Processor (ISP). This approach significantly reduces hardware requirements and memory consumption, which is crucial for robotic vision systems. Our method introduces two custom-designed convolutional kernels capable of performing convolutions directly on raw images, preserving inter-channel information without converting to RGB. Experimental results show that our network outperforms existing algorithms on raw images, achieving higher accuracy and stability under large rotations and scale variations. This work represents the first attempt to develop a keypoint detection and feature description network specifically for raw images, offering a more efficient solution for resource-constrained environments.