AXERA-TECH / ONNX-YOLO-World-Open-Vocabulary-Object-DetectionLinks
Python scripts performing Open Vocabulary Object Detection using the YOLO-World model in ONNX. And Export the ONNX model for AXera's NPU
☆12Updated last month
Alternatives and similar repositories for ONNX-YOLO-World-Open-Vocabulary-Object-Detection
Users that are interested in ONNX-YOLO-World-Open-Vocabulary-Object-Detection are comparing it to the libraries listed below
Sorting:
- FastSAM 部署版本,便于移植不同平,部署简单、运行速度快。☆23Updated last year
- DETR tensor去除推理过程无用辅助头+fp16部署再次加速+解决转tensorrt 输出全为0问题的新方法。☆12Updated last year
- c++实现的clip推理,模型有一点点改动,但是不大,改动和导出模型的代码可以在readme里找到,模型文件都在Releases里,包括AX650的模型。新增支持ChineseCLIP☆30Updated 3 months ago
- The rknn2 API uses the secondary encapsulation of the process, which is easy for everyone to call. It is applicable to rk356x rk3588☆46Updated 3 years ago
- 使用NCNN推理框架和ByteTrack目标跟踪框架,对网络、文件流URL进行实时性视频推理,而UI界面则由Qt框架实现☆21Updated 11 months ago
- ☆27Updated 3 months ago
- Quantize yolov5 using pytorch_quantization.🚀🚀🚀☆14Updated last year
- ☆23Updated last year
- ☆39Updated 11 months ago
- 使用ONNXRuntime部署PP-YOLOE目标检测,支持PP-YOLOE-s、PP-YOLOE-m、PP-YOLOE-l、PP-YOLOE-x四种结构,包含C++和Python两个版本的程序☆21Updated 3 years ago
- ncnn 实现一些项目例子☆26Updated 2 years ago
- 对 tensorRT_Pro 开源项目理解☆21Updated 2 years ago
- ☆17Updated last year
- ppstructure deploy by ncnn☆33Updated last year
- 这是一个使用opencv读取视频并使用socket进行传输视频画面的脚本文件,相较于调用ffmpeg传输节约了90%的数据量☆11Updated last year
- ☆24Updated 2 years ago
- ffmpeg+cuvid+tensorrt+multicamera☆12Updated 9 months ago
- SAM and lama inpaint,包含QT的GUI交互界面,实现了交互式可实时显示结果的画点、画框进行SAM,然后通过进行Inpaint,具体操作看readme里的视频。☆50Updated last year
- A unified and extensible pipeline for deep learning model inference with C++. Now support yolov8, yolov9, clip, and nanosam. More models …☆12Updated 2 months ago
- ☆67Updated 3 years ago
- RKNN模型推理部署模板☆23Updated 2 years ago
- ☆28Updated 3 years ago
- Nanodet with mosaic and mixup☆13Updated 4 years ago
- yolov8seg 瑞芯微 rknn 板端 C++部署,使用平台 rk3588。☆28Updated last year
- 使用ONNXRuntime部署DeDoDe:"局部特征匹配:检测,不要描述——描述,不要检测"。依然是C++和Python两个版本的程序☆23Updated last year
- A Python and C++ library for model encryption and decryption, built on Crypto++, with support for various deep learning frameworks includ…☆48Updated last year
- ☆16Updated last year
- naive c++ version of Mask2Former with ncnn☆23Updated 3 years ago
- 高效部署:YOLO X, V3, V4, V5, V6, V7, V8, EdgeYOLO TRT推理 ™️ ,前后处理均由CUDA核函数实现 CPP/CUDA🚀☆50Updated 2 years ago
- Try to export the ONNX QDQ model that conforms to the AXERA NPU quantization specification. Currently, only w8a8 is supported.☆11Updated last year