Dynamic Neural Field Modeling of Visual Contrast for Perceiving Incoherent Looming

2025-04-06

AI Summary

This work addresses the poor robustness of looming target detection in complex dynamic scenes—such as rainy or foggy conditions—where visual cues become inconsistent. To tackle this, we propose a novel dual-channel Dynamic Neural Field (DNF) model, termed the ON/OFF Contrast Field, which is the first to integrate the biologically inspired ON/OFF contrast encoding mechanism into the DNF framework. The model employs Gaussian-normalized lateral excitatory kernels and a multi-field coupled integration scheme, enabling synergistic fusion of dual-channel responses within a summation field to enhance selective sensitivity to incoherent, noise-corrupted looming stimuli. Experiments on synthetic rain- and fog-degraded scenes demonstrate that our method achieves over 35% higher accuracy than state-of-the-art locust-inspired models, while maintaining low computational power consumption and strong environmental robustness. This establishes a new paradigm for collision warning systems operating in realistic, challenging environments.

Technology Category

• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Computer Vision: Adversarial Attacks & Robustness
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems

Abstract

Amari's Dynamic Neural Field (DNF) framework provides a brain-inspired approach to modeling the average activation of neuronal groups. Leveraging a single field, DNF has become a promising foundation for low-energy looming perception module in robotic applications. However, the previous DNF methods face significant challenges in detecting incoherent or inconsistent looming features--conditions commonly encountered in real-world scenarios, such as collision detection in rainy weather. Insights from the visual systems of fruit flies and locusts reveal encoding ON/OFF visual contrast plays a critical role in enhancing looming selectivity. Additionally, lateral excitation mechanism potentially refines the responses of loom-sensitive neurons to both coherent and incoherent stimuli. Together, these offer valuable guidance for improving looming perception models. Building on these biological evidence, we extend the previous single-field DNF framework by incorporating the modeling of ON/OFF visual contrast, each governed by a dedicated DNF. Lateral excitation within each ON/OFF-contrast field is formulated using a normalized Gaussian kernel, and their outputs are integrated in the Summation field to generate collision alerts. Experimental evaluations show that the proposed model effectively addresses incoherent looming detection challenges and significantly outperforms state-of-the-art locust-inspired models. It demonstrates robust performance across diverse stimuli, including synthetic rain effects, underscoring its potential for reliable looming perception in complex, noisy environments with inconsistent visual cues.

Problem

• Detects incoherent looming features in real-world scenarios
• Incorporates ON/OFF visual contrast modeling for enhanced selectivity
• Improves collision detection in noisy, inconsistent visual environments

Innovation

• Incorporates ON/OFF visual contrast modeling
• Uses lateral excitation with Gaussian kernel
• Integrates outputs in Summation field