PixRO: Pixel-Distributed Rotational Odometry with Gaussian Belief Propagation

2024-06-14

arXiv.org

AI Summary

Traditional vision algorithms process full-frame inputs, incurring pixel-level redundancy and noise that exacerbate transmission energy consumption and computational overhead. To address this, we propose Pixel-level Distributed Rotational Odometry (PDRO), the first method to decompose inter-frame rotational estimation at the single-pixel level. Each pixel independently leverages only its local gradient and messages exchanged with neighboring pixels, enabling decentralized, low-redundancy global motion estimation via Gaussian Belief Propagation (GBP). PDRO integrates pixel-wise local modeling, distributed message passing, and lightweight visual inference to substantially suppress irrelevant information. We validate PDRO’s accuracy and efficiency across multiple real-world datasets. The code is publicly released. By delivering compact, information-rich motion cues—rather than raw pixel streams—PDRO serves as an efficient front-end for downstream perception tasks.

Technology Category

• Computer Vision: Low Level & Physics-based Vision
• Search and Optimization: Distributed Search
• Intelligent Robots: State Estimation

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Experiences and lessons learnt from Web-based algorithms and system deployments
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs

Abstract

Visual sensors are not only becoming better at capturing high-quality images but also they have steadily increased their capabilities in processing data on their own on-chip. Yet the majority of VO pipelines rely on the transmission and processing of full images in a centralized unit (e.g. CPU or GPU), which often contain much redundant and low-quality information for the task. In this paper, we address the task of frame-to-frame rotational estimation but, instead of reasoning about relative motion between frames using the full images, distribute the estimation at pixel-level. In this paradigm, each pixel produces an estimate of the global motion by only relying on local information and local message-passing with neighbouring pixels. The resulting per-pixel estimates can then be communicated to downstream tasks, yielding higher-level, informative cues instead of the original raw pixel-readings. We evaluate the proposed approach on real public datasets, where we offer detailed insights about this novel technique and open-source our implementation for the future benefit of the community.

Problem

• Distributing rotational odometry computation across image pixels
• Reducing computational overhead by processing vision in-pixel
• Achieving global motion consensus through pixel-level messaging

Innovation

• Distributed rotational odometry at pixel level
• Gaussian Belief Propagation for decentralized inference
• Pixel-level global motion estimation via messaging-passing