Pinned Loading

1. YOLOv5-ncnn-OpenVINO-MNN-ONNXRuntime-OpenCV-CPP YOLOv5-ncnn-OpenVINO-MNN-ONNXRuntime-OpenCV-CPP Public

   YOLOv5 C++ inference implemented using multiple frameworks: ncnn, OpenVINO, MNN, ONNXRuntime, and OpenCV.

   C++ 18 2

2. optimize-gemm optimize-gemm Public

   My gemm optimization on RPi (ARM) achieved a 170x performance boost, showing speeds faster than Eigen and close to OpenBLAS.

   C++ 15

3. tiny-cnn tiny-cnn Public

   A tiny CNN which is extremely fast and lightweight, beat ONNXRuntime and ncnn.

   C 7

4. PaddleOCR-ncnn-CPP PaddleOCR-ncnn-CPP Public

   C++ implementations of PP-OCRv3 and PP-OCRv5 using ncnn for inference.

   C++ 51 3

5. Peterande/D-FINE Peterande/D-FINE Public

   D-FINE: Redefine Regression Task of DETRs as Fine-grained Distribution Refinement [ICLR 2025 Spotlight]

   Python 3k 276

6. Handcrafted-CNN Handcrafted-CNN Public

   A handcrafted convolutional neural network. Matrix multiplication instead of loop operation, very fast.

   Python 3