Detect from Prompt with C++ and TensorRT
A high-performance, open-vocabulary object detector in C++ accelerated with TensorRT.
GitHub Link: https://github.com/sanket-pixel/YOLO-World-TensorRT-CPP
This project is a high-performance C++ implementation of the YOLO-World model for real-time, open-vocabulary object detection. The entire inference pipeline is accelerated using NVIDIA TensorRT and CUDA, allowing the system to detect any object in an image based on a natural language text prompt.
The application bypasses Python dependencies for a pure C++ backend, making it suitable for production and edge deployment scenarios where speed and efficiency are critical.