Python acts as the orchestration layer, while a native C bridge launches the RISC-V executable through QEMU and returns results back to Python. The project is structured to demonstrate low-level ...
The GPU implementation uses custom CUDA kernels. The blur and Sobel stages use shared-memory tiling to reduce repeated global-memory access for neighborhood operations. This project demonstrates ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results