fergalriordan/mamba-rs-engine

High-throughput C++/CUDA inference engine for Satellite Change Detection using RS-Mamba. Optimised for edge-deployment in environmental and humanitarian monitoring. # Mamba-RS-Engine High-throughput inference engine for Remote Sensing (RS) change detection, utilizing **State Space Models (RS-Mamba)** and optimized via **C++/TensorRT** for edge-constrained environments. ## Project Goals - **Architecture:** Implement an omnidirectional State Space Model (RS-Mamba) to bypass the $O(N^2)$ complexity of Vision Transformers in high-res satellite imagery. - **Domain:** Rapid identification of environmental changes (Wildfires/Flooding) using the LEVIR-CD or OSCD datasets. - **Optimisation:** Develop a standalone C++ inference server achieving >2x throughput vs. PyTorch baselines. --- *Developed by Fergal Riordan as part of ongoing upskilling in MLE focusing on Physical AI and Environmental Tech.*