Visual Odometry Feature Restoration via Geometrical Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Dynamic Vision Sensor (DVS) feature tracking fails due to inconsistent features across adjacent frames, leading to challenges in visual odometry applications.
Innovation Solution
A method and system for analyzing images captured from a camera, which involves receiving images, performing feature extraction and tracking, detecting untracked features, and executing a geometrical prediction to predict tracked features from untracked ones.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If DVS feature tracking is performed using optical flow methods, then tracking speed is maintained, but tracking reliability deteriorates due to inconsistent DVS features across adjacent frames
Solution Approach 1:
The patent introduces an intermediary restoration module that bridges the gap between DVS feature extraction and optical flow tracking. This module uses geometric prediction and outlier removal to restore consistent features from inconsistent DVS data, enabling reliable tracking without increasing overall system complexity. The intermediary processes transform unreliable DVS features into reliable inputs for optical flow algorithms.
2Stability of the object's composition
If geometric prediction is applied to restore untracked features, then feature consistency improves, but computational time increases
Solution Approach 1:
The patent applies geometric prediction selectively rather than to all features. The outlier removal module identifies and restores only the untracked features that contribute to inconsistency, leaving already-tracked features unchanged. This partial action approach maintains feature consistency while minimizing unnecessary computational overhead and processing time.
3Measurement precision
If outlier removal is performed to eliminate motion-dependent information, then measurement precision improves, but processing complexity increases
Solution Approach 1:
The patent performs outlier removal as a preliminary step before feature tracking and restoration. By pre-processing the DVS features to eliminate motion-dependent outliers early in the pipeline, the system reduces the burden on subsequent tracking algorithms. This preliminary action improves measurement precision while keeping the overall processing complexity manageable through staged processing.
Data Source
AI summary
The present disclosure provides a method of analyzing an image captured from an imaging device. The method includes receiving one or more images captured from a camera over a predetermined time period. Feature extraction and tracking are executed with respect to the received images. One or more features are detected as untracked features during the feature extraction and tracking. A geometrical prediction is executed with respect to the untracked features to predict one or more features and thereby achieving the one or more untracked features as the one or more tracked features.