Optimization and Implementation of Synthetic Basis Feature Descriptor on FPGA

Abstract

Feature detection, description, and matching are crucial steps for many computer vision algorithms. These steps rely on feature descriptors to match image features across sets of images. Previous work has shown that our SYnthetic BAsis (SYBA) feature descriptor can offer superior performance to other binary descriptors. This paper focused on various optimizations and hardware implementation of the newer and optimized version. The hardware implementation on a field-programmable gate array (FPGA) is a high-throughput low-latency solution which is critical for applications such as high-speed object detection and tracking, stereo vision, visual odometry, structure from motion, and optical flow. We compared our solution to other hardware designs of binary descriptors. We demonstrated that our implementation of SYBA as a feature descriptor in hardware offered superior image feature matching performance and used fewer resources than most binary feature descriptor implementations.

Ego-Motion Estimation Using Recurrent Convolutional Neural Networks through Optical Flow Learning

“…Geometry-based VO methods can be divided into feature-based methods and direct methods. Feature-based methods estimate motion based on geometric constraints extracted from imagery [7][8][9], while direct methods optimize the photometric error of the whole image or local area to estimate motion. Specifically, the feature-based methods detect and track a set of sparse salient features between consecutive frames and then calculate the pose parameters by analyzing the position changes of the feature points in the consecutive images.…”

“…Classic geometry-based VO approaches rely on the geometric constraints extracted from imagery for pose estimation. They typically consist of a complicated pipeline including camera calibration, feature detection, feature matching (or tracking), outlier rejection (e.g., RANSAC), motion estimation, scale estimation, and local optimization (Bundle Adjustment) [7][8][9]. In virtue of Convolutional Neural Network (CNN) representational power, learning-based VO in the last few years has seen increasing attention and achieved promising progress because of its desirable properties of robustness to image noise and camera calibration independence.…”