Vector-Symbolic Architecture for Event-Based Optical Flow

2024-05-14

arXiv.org

AI Summary

Event-based optical flow estimation suffers from insufficient robustness in feature matching due to the absence of grayscale intensity information. Method: This work pioneers the integration of Vector Symbolic Architecture (VSA) into event-based flow estimation, proposing a high-dimensional, polarity-aware, multi-scale binary/real-valued event feature encoding scheme. We design a dual-path framework: (i) VSA-Flow, a model-driven approach leveraging VSA’s topological similarity for stable inter-frame correspondence; and (ii) VSA-SM, the first end-to-end self-supervised method relying solely on events, optimized via similarity maximization. Contribution/Results: Our VSA-based representations ensure geometrically consistent matching across frames without grayscale cues. On DSEC, our method significantly outperforms existing model-driven and self-supervised approaches; on MVSEC, it maintains state-of-the-art performance—demonstrating full independence from grayscale images.

Technology Category

• Computer Vision: Visual Reasoning & Symbolic Representations
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Machine Learning: Neuro-Symbolic Learning

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Web query analysis, representation and understanding
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs

Abstract

From a perspective of feature matching, optical flow estimation for event cameras involves identifying event correspondences by comparing feature similarity across accompanying event frames. In this work, we introduces an effective and robust high-dimensional (HD) feature descriptor for event frames, utilizing Vector Symbolic Architectures (VSA). The topological similarity among neighboring variables within VSA contributes to the enhanced representation similarity of feature descriptors for flow-matching points, while its structured symbolic representation capacity facilitates feature fusion from both event polarities and multiple spatial scales. Based on this HD feature descriptor, we propose a novel feature matching framework for event-based optical flow, encompassing both model-based (VSA-Flow) and self-supervised learning (VSA-SM) methods. In VSA-Flow, accurate optical flow estimation validates the effectiveness of HD feature descriptors. In VSA-SM, a novel similarity maximization method based on the HD feature descriptor is proposed to learn optical flow in a self-supervised way from events alone, eliminating the need for auxiliary grayscale images. Evaluation results demonstrate that our VSA-based method achieves superior accuracy in comparison to both model-based and self-supervised learning methods on the DSEC benchmark, while remains competitive among both methods on the MVSEC benchmark. This contribution marks a significant advancement in event-based optical flow within the feature matching methodology.

Problem

• Event Camera
• Optical Flow Estimation
• Action-based Light Variation

Innovation

• Vector Symbolic Architecture
• Event Camera Optical Flow
• Self-Learning Matching