AI model deployment based on NVIDIA and Qualcomm platforms
The repository mainly provides 2D model inference functionality, and the code provides daily development of packaged libs for integration, testing, and inference. The framework provides multi-threaded, singleton pattern, producer and consumer patterns. Cache log analysis is also supported.
CVDeploy-2D Features:
Supports real-time caching of runtime log files
Offers easy integration via a singleton design pattern, providing header files and libraries to encapsulate the algorithm program
Supports YAML configuration files to set essential parameters for model files
Compatible with various hardware platforms, including NVIDIA and Qualcomm
Supports multiple 2D models, such as YOLOv5 and YOLOX
Integrates multithreading with producer-consumer mode to enable concurrent processing
Includes user-friendly Bash scripts for one-click installation and execution
Planned Features for CVDeploy-2D:
Support for model compression techniques, such as quantization.
Memory leak detection and precision validation for ONNX, TRT, and QNN models.
Integration of dynamic object detection with geometric tracking algorithms.
Documentation#
Getting Started
Framework