The field of neuromorphic computing and event-based vision is rapidly advancing, with a focus on developing more efficient and scalable solutions for real-time processing and learning. Recent research has explored the use of spiking neural networks (SNNs) and event-based cameras for applications such as object detection, optical flow estimation, and robotic perception. Notable developments include the introduction of novel learning paradigms, such as Spike Agreement Dependent Plasticity, and the demonstration of biologically realistic simulations of brain connectomes on neuromorphic hardware. Furthermore, researchers have made significant progress in improving the energy efficiency and performance of SNNs, with some studies achieving state-of-the-art results in visual detection tasks with ultra-low latency. The use of event-based cameras has also been explored for applications such as robotic navigation and manipulation, with promising results. Overall, the field is moving towards more efficient, scalable, and biologically plausible solutions for real-time processing and learning. Noteworthy papers include the introduction of TaiBai, a fully programmable brain-inspired processor, and the demonstration of EventTracer, a path tracing-based rendering pipeline for simulating high-fidelity event sequences.
Advances in Neuromorphic Computing and Event-Based Vision
Sources
Energy-Information Trade-Off in Self-Directed Channel Memristors
Spike Agreement Dependent Plasticity: A scalable Bio-Inspired learning paradigm for Spiking Neural Networks
Neuromorphic Simulation of Drosophila Melanogaster Brain Connectome on Loihi 2
Reinforcement-Guided Hyper-Heuristic Hyperparameter Optimization for Fair and Explainable Spiking Neural Network-Based Financial Fraud Detection
SEBVS: Synthetic Event-based Visual Servoing for Robot Navigation and Manipulation
AQ-PCDSys: An Adaptive Quantized Planetary Crater Detection System for Autonomous Space Exploration
EventTracer: Fast Path Tracing-based Event Stream Rendering
TaiBai: A fully programmable brain-inspired processor with topology-aware efficiency
VibES: Induced Vibration for Persistent Event-Based Sensing
When Routers, Switches and Interconnects Compute: A processing-in-interconnect Paradigm for Scalable Neuromorphic AI
Hardware-aware vs. Hardware-agnostic Energy Estimation for SNN in Space Applications
Context-aware Sparse Spatiotemporal Learning for Event-based Vision
Task-Aware Tuning of Time Constants in Spiking Neural Networks for Multimodal Classification
Improving Liver Disease Diagnosis with SNNDeep: A Custom Spiking Neural Network Using Diverse Learning Algorithms
Ultra-Low-Latency Spiking Neural Networks with Temporal-Dependent Integrate-and-Fire Neuron Model for Objects Detection
Encoding Tactile Stimuli for Organoid Intelligence in Braille Recognition