{"id":13435952,"url":"https://github.com/laugh12321/TensorRT-YOLO","last_synced_at":"2025-03-18T12:30:47.715Z","repository":{"id":219539656,"uuid":"749292767","full_name":"laugh12321/TensorRT-YOLO","owner":"laugh12321","description":"🚀 你的YOLO部署神器。TensorRT Plugin、CUDA Kernel、CUDA Graphs三管齐下，享受闪电般的推理速度。| Your YOLO Deployment Powerhouse. With the synergy of TensorRT Plugins, CUDA Kernels, and CUDA Graphs, experience lightning-fast inference speeds.","archived":false,"fork":false,"pushed_at":"2024-10-11T07:38:56.000Z","size":4560,"stargazers_count":555,"open_issues_count":2,"forks_count":67,"subscribers_count":5,"default_branch":"main","last_synced_at":"2024-10-13T19:47:43.241Z","etag":null,"topics":["cuda","cuda-graph","cuda-kernels","cuda-programming","detection","onnx","ppyoloe","tensorrt","yolov10","yolov3","yolov5","yolov6","yolov7","yolov8","yolov9"],"latest_commit_sha":null,"homepage":"https://github.com/laugh12321/TensorRT-YOLO","language":"C++","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"gpl-3.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/laugh12321.png","metadata":{"files":{"readme":"README.en.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2024-01-28T06:10:50.000Z","updated_at":"2024-10-11T07:49:58.000Z","dependencies_parsed_at":"2024-03-31T16:23:45.424Z","dependency_job_id":"8b27666e-2a25-4318-9dfd-11891117e3da","html_url":"https://github.com/laugh12321/TensorRT-YOLO","commit_stats":null,"previous_names":["laugh12321/tensorrt-yolo"],"tags_count":10,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/laugh12321%2FTensorRT-YOLO","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/laugh12321%2FTensorRT-YOLO/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/laugh12321%2FTensorRT-YOLO/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/laugh12321%2FTensorRT-YOLO/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/laugh12321","download_url":"https://codeload.github.com/laugh12321/TensorRT-YOLO/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":244222110,"owners_count":20418447,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["cuda","cuda-graph","cuda-kernels","cuda-programming","detection","onnx","ppyoloe","tensorrt","yolov10","yolov3","yolov5","yolov6","yolov7","yolov8","yolov9"],"created_at":"2024-07-31T03:00:41.260Z","updated_at":"2025-03-18T12:30:47.708Z","avatar_url":"https://github.com/laugh12321.png","language":"C++","funding_links":[],"categories":["Lighter and Deployment Frameworks","Applications","C++"],"sub_categories":[],"readme":"English | [简体中文](README.md)\n\n\u003cdiv align=\"center\"\u003e\n  \u003cimg width=\"75%\" src=\"assets/logo.png\"\u003e\n  \n  \u003cp align=\"center\"\u003e\n      \u003ca href=\"./LICENSE\"\u003e\u003cimg alt=\"GitHub License\" src=\"https://img.shields.io/github/license/laugh12321/TensorRT-YOLO?style=for-the-badge\u0026color=0074d9\"\u003e\u003c/a\u003e\n      \u003ca href=\"https://github.com/laugh12321/TensorRT-YOLO/releases\"\u003e\u003cimg alt=\"GitHub Release\" src=\"https://img.shields.io/github/v/release/laugh12321/TensorRT-YOLO?style=for-the-badge\u0026color=0074d9\"\u003e\u003c/a\u003e\n      \u003cimg alt=\"GitHub Repo Stars\" src=\"https://img.shields.io/github/stars/laugh12321/TensorRT-YOLO?style=for-the-badge\u0026color=3dd3ff\"\u003e\n      \u003cimg alt=\"Linux\" src=\"https://img.shields.io/badge/Linux-FCC624?style=for-the-badge\u0026logo=linux\u0026logoColor=black\"\u003e\n      \u003cimg alt=\"Arch\" src=\"https://img.shields.io/badge/Arch-x86%20%7C%20ARM-0091BD?style=for-the-badge\u0026logo=cpu\u0026logoColor=white\"\u003e\n      \u003cimg alt=\"NVIDIA\" src=\"https://img.shields.io/badge/NVIDIA-%2376B900.svg?style=for-the-badge\u0026logo=nvidia\u0026logoColor=white\"\u003e\n  \u003c/p\u003e\n\n  \u003cp align=\"center\"\u003e\n      \u003ca href=\"/docs/en/build_and_install.md\"\u003e\u003cimg src=\"https://img.shields.io/badge/-Installation-0078D4?style=for-the-badge\u0026logo=github\u0026logoColor=white\"\u003e\u003c/a\u003e\n      \u003ca href=\"/examples/\"\u003e\u003cimg src=\"https://img.shields.io/badge/-Usage Examples-0078D4?style=for-the-badge\u0026logo=github\u0026logoColor=white\"\u003e\u003c/a\u003e\n      \u003ca href=\"#quick-start\"\u003e\u003cimg src=\"https://img.shields.io/badge/-Quick Start-0078D4?style=for-the-badge\u0026logo=github\u0026logoColor=white\"\u003e\u003c/a\u003e\n      \u003ca href=\"\"\u003e\u003cimg src=\"https://img.shields.io/badge/-API Documentation-0078D4?style=for-the-badge\u0026logo=github\u0026logoColor=white\"\u003e\u003c/a\u003e\n      \u003ca href=\"https://github.com/laugh12321/TensorRT-YOLO/releases\"\u003e\u003cimg src=\"https://img.shields.io/badge/-Release Notes-0078D4?style=for-the-badge\u0026logo=github\u0026logoColor=white\"\u003e\u003c/a\u003e\n  \u003c/p\u003e\n\n\u003c/div\u003e\n\n---\n\n🚀 TensorRT-YOLO is an **easy-to-use**, **extremely efficient** inference deployment tool for the **YOLO series** designed specifically for NVIDIA devices. The project not only integrates TensorRT plugins to enhance post-processing but also utilizes CUDA kernels and CUDA graphs to accelerate inference. TensorRT-YOLO provides support for both C++ and Python inference, aiming to deliver a 📦**out-of-the-box** deployment experience. It covers various task scenarios such as [object detection](examples/detect/), [instance segmentation](examples/segment/), [image classification](examples/classify/), [pose estimation](examples/pose/), [oriented object detection](examples/obb/), and [video analysis](examples/VideoPipe), meeting developers' deployment needs in **multiple scenarios**.\n\n\u003cdiv align=\"center\"\u003e\n\n[\u003cimg src='assets/obb.png' height=\"138px\" width=\"190px\"\u003e](examples/obb/)\n[\u003cimg src='assets/detect.jpg' height=\"138px\" width=\"190px\"\u003e](examples/detect/)\n[\u003cimg src='assets/segment.jpg' height=\"138px\" width=\"190px\"\u003e](examples/segment/)\n[\u003cimg src='assets/pose.jpg' height=\"138px\" width=\"190px\"\u003e](examples/pose/)\n[\u003cimg src='assets/example.gif' width=\"770px\"\u003e](examples/videopipe)\n\n\u003c/div\u003e\n\n## \u003cdiv align=\"center\"\u003e🌠 Recent updates\u003c/div\u003e\n\n- [Performance Leap! TensorRT-YOLO 6.0: Comprehensive Upgrade Analysis and Practical Guide](https://medium.com/@laugh12321/performance-leap-tensorrt-yolo-6-0-comprehensive-upgrade-analysis-and-practical-guide-9d19ad3b53f9) 🌟 NEW\n\n## \u003cdiv align=\"center\"\u003e✨ Key Features\u003c/div\u003e\n\n### 🎯 Diverse YOLO Support\n- **Comprehensive Compatibility**: Supports YOLOv3 to YOLOv11 series models, as well as PP-YOLOE and PP-YOLOE+, meeting diverse needs.\n- **Flexible Switching**: Provides simple and easy-to-use interfaces for quick switching between different YOLO versions. 🌟 NEW\n- **Multi-Scenario Applications**: Offers rich example codes covering [Detect](examples/detect/), [Segment](examples/segment/), [Classify](examples/classify/), [Pose](examples/pose/), [OBB](examples/obb/), and more.\n\n### 🚀 Performance Optimization\n- **CUDA Acceleration**: Optimizes pre-processing through CUDA kernels and accelerates inference using CUDA graphs.\n- **TensorRT Integration**: Deeply integrates TensorRT plugins to significantly speed up post-processing and improve overall inference efficiency.\n- **Multi-Context Inference**: Supports multi-context parallel inference to maximize hardware resource utilization. 🌟 NEW\n- **Memory Management Optimization**: Adapts multi-architecture memory optimization strategies (e.g., Zero Copy mode for Jetson) to enhance memory efficiency. 🌟 NEW\n\n### 🛠️ Usability\n- **Out-of-the-Box**: Provides comprehensive C++ and Python inference support to meet different developers' needs.\n- **CLI Tools**: Built-in command-line tools for quick model export and inference, improving development efficiency.\n- **Docker Support**: Offers one-click Docker deployment solutions to simplify environment configuration and deployment processes.\n- **No Third-Party Dependencies**: All functionalities are implemented using standard libraries, eliminating the need for additional dependencies and simplifying deployment.\n- **Easy Deployment**: Provides dynamic library compilation support for easy calling and deployment.\n\n### 🌐 Compatibility\n- **Multi-Platform Support**: Fully compatible with various operating systems and hardware platforms, including Windows, Linux, ARM, and x86.\n- **TensorRT Compatibility**: Perfectly adapts to TensorRT 10.x versions, ensuring seamless integration with the latest technology ecosystem.\n\n### 🔧 Flexible Configuration\n- **Customizable Preprocessing Parameters**: Supports flexible configuration of various preprocessing parameters, including **channel swapping (SwapRB)**, **normalization parameters**, and **border padding**. 🌟 NEW\n\n## \u003cdiv align=\"center\"\u003e🚀 Performance\u003c/div\u003e\n\n\u003cdiv align=\"center\"\u003e\n\n| Model | Official + trtexec (ms) | trtyolo + trtexec (ms) | TensorRT-YOLO Inference (ms)|\n|:-----:|:-----------------------:|:----------------------:|:---------------------------:|\n| YOLOv11n | 1.611 ± 0.061        | 1.428 ± 0.097          | 1.228 ± 0.048               |\n| YOLOv11s | 2.055 ± 0.147        | 1.886 ± 0.145          | 1.687 ± 0.047               |\n| YOLOv11m | 3.028 ± 0.167        | 2.865 ± 0.235          | 2.691 ± 0.085               |\n| YOLOv11l | 3.856 ± 0.287        | 3.682 ± 0.309          | 3.571 ± 0.102               |\n| YOLOv11x | 6.377 ± 0.487        | 6.195 ± 0.482          | 6.207 ± 0.231               |\n\n\u003c/div\u003e\n\n\u003e [!NOTE]\n\u003e\n\u003e **Testing Environment**\n\u003e - **GPU**: NVIDIA RTX 2080 Ti 22GB\n\u003e - **Input Size**: 640×640 pixels\n\u003e\n\u003e **Testing Tools**\n\u003e - **Official**: Using the ONNX model exported by Ultralytics.\n\u003e - **trtyolo**: Using the CLI tool (trtyolo) provided by TensorRT-YOLO to export the ONNX model with the EfficientNMS plugin.\n\u003e - **trtexec**: Using NVIDIA's `trtexec` tool to build the ONNX model into an engine and perform inference testing.\n\u003e   - **Build Command**: `trtexec --onnx=xxx.onnx --saveEngine=xxx.engine --fp16`\n\u003e   - **Test Command**: `trtexec --avgRuns=1000 --useSpinWait --loadEngine=xxx.engine`\n\u003e - **TensorRT-YOLO Inference**: Using the TensorRT-YOLO framework to measure the latency (including pre-processing, inference, and post-processing) of the engine obtained through the **trtyolo + trtexec** method.\n\n## \u003cdiv align=\"center\"\u003e🔮 Documentation\u003c/div\u003e\n\n- **Installation Guide**\n    - [📦 Quick Compilation and Installation](docs/en/build_and_install.md)\n- **Usage Examples**\n    - [Object Detection Example](examples/detect/README.en.md)\n    - [Instance Segmentation Example](examples/segment/README.en.md)\n    - [Image Classification Example](examples/classify/README.en.md)\n    - [Pose Estimation Example](examples/pose/README.en.md)\n    - [Oriented Object Detection Example](examples/obb/README.en.md)\n    - [📹 Video Analysis Example](examples/VideoPipe/README.en.md)\n    - [Multi-threading and Multi-processing Example](examples/mutli_thread/README.en.md) 🌟 NEW\n- **API Documentation**\n    - Python API Documentation (⚠️ Not Implemented)\n    - C++ API Documentation (⚠️ Not Implemented)\n- **FAQ**\n    - ⚠️ Collecting ...\n- **Supported Models List**\n    - [🖥️ Supported Models List](#support-models)\n\n## \u003cdiv align=\"center\"\u003e💨 Quick Start\u003c/div\u003e\u003cdiv id=\"quick-start\"\u003e\u003c/div\u003e\n\n### 1. Prerequisites\n\n- **CUDA**: Recommended version ≥ 11.0.1\n- **TensorRT**: Recommended version ≥ 8.6.1\n- **Operating System**: Linux (x86_64 or arm) (recommended); Windows is also supported\n\n### 2. Installation\n\n- Refer to the [📦 Quick Compilation and Installation](docs/en/build_and_install.md) documentation.\n\n### 3. Model Export\n\n- Refer to the [🔧 Model Export](docs/en/model_export.md) documentation to export an ONNX model suitable for inference in this project and build it into a TensorRT engine.\n\n### 4. Inference Example\n\n\u003e [!NOTE]\n\u003e\n\u003e `ClassifyModel`, `DetectModel`, `OBBModel`, `SegmentModel`, and `PoseModel` correspond to image classification (Classify), detection (Detect), oriented bounding box (OBB), segmentation (Segment), and pose estimation (Pose) models, respectively.\n\n- Inference using Python:\n\n  ```python\n  import cv2\n  from tensorrtyolo.infer import InferOption, DetectModel, generatelabels, visualize\n  \n  def main():\n      # -------------------- Initialization --------------------\n      # Configure inference settings\n      option = InferOption()\n      option.enableswaprb()  # Convert OpenCV's default BGR format to RGB\n      # Special model configuration example (uncomment for PP-YOLOE series)\n      # option.setnormalizeparams([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])\n  \n      # -------------------- Model Initialization --------------------\n      # Load TensorRT engine file (ensure the path is correct)\n      # Note: Initial engine loading may take longer due to optimization\n      model = DetectModel(engine_path=\"yolo11n-with-plugin.engine\", \n                        option=option)\n  \n      # -------------------- Data Preprocessing --------------------\n      # Load test image (add file existence check)\n      inputimg = cv2.imread(\"testimage.jpg\")\n      if input_img is None:\n          raise FileNotFoundError(\"Failed to load test image. Check the file path.\")\n  \n      # -------------------- Inference Execution --------------------\n      # Perform object detection (returns bounding boxes, confidence scores, and class labels)\n      detectionresult = model.predict(inputimg)\n      print(f\"==\u003e Detection Result: {detection_result}\")\n  \n      # -------------------- Result Visualization --------------------\n      # Load class labels (ensure labels.txt matches the model)\n      classlabels = generate_labels(labelsfile=\"labels.txt\")\n      # Generate visualized result\n      visualized_img = visualize(\n          image=input_img,\n          result=detection_result,\n          labels=class_labels,\n      )\n      cv2.imwrite(\"visimage.jpg\", visualizedimg)\n  \n      # -------------------- Model Cloning Demo --------------------\n      # Clone model instance (for multi-threaded scenarios)\n      cloned_model = model.clone()  # Create an independent copy to avoid resource contention\n      # Verify cloned model inference consistency\n      clonedresult = cloned_model.predict(inputimg)\n      print(f\"==\u003e Cloned Result: {cloned_result}\")\n  \n  if _name__ == \"__main_\":\n      main()\n  ```\n\n- Inference using C++:\n\n  ```cpp\n  #include \u003cmemory\u003e\n  #include \u003copencv2/opencv.hpp\u003e\n  \n  // For ease of use, the module uses only CUDA and TensorRT, with the rest implemented in standard libraries\n  #include \"deploy/model.hpp\"  // Contains model inference class definitions\n  #include \"deploy/option.hpp\"  // Contains inference option configuration class definitions\n  #include \"deploy/result.hpp\"  // Contains inference result definitions\n  \n  int main() {\n      try {\n          // -------------------- Initialization --------------------\n          deploy::InferOption option;\n          option.enableSwapRB();  // BGR-\u003eRGB conversion\n          \n          // Special model parameter setup example\n          // const std::vector\u003cfloat\u003e mean{0.485f, 0.456f, 0.406f};\n          // const std::vector\u003cfloat\u003e std{0.229f, 0.224f, 0.225f};\n          // option.setNormalizeParams(mean, std);\n  \n          // -------------------- Model Initialization --------------------\n          auto detector = std::make_unique\u003cdeploy::DetectModel\u003e(\n              \"yolo11n-with-plugin.engine\",  // Model path\n              option                         // Inference settings\n          );\n  \n          // -------------------- Data Loading --------------------\n          cv::Mat cvimage = cv::imread(\"testimage.jpg\");\n          if (cv_image.empty()) {\n              throw std::runtime_error(\"Failed to load test image.\");\n          }\n          \n          // Encapsulate image data (no pixel data copying)\n          deploy::Image input_image(\n              cv_image.data,     // Pixel data pointer\n              cv_image.cols,     // Image width\n              cv_image.rows,     // Image height\n          );\n  \n          // -------------------- Inference Execution --------------------\n          deploy::DetResult result = detector-\u003epredict(input_image);\n          std::cout \u003c\u003c result \u003c\u003c std::endl;\n  \n          // -------------------- Result Visualization (Example) --------------------\n          // Implement visualization logic in actual development, e.g.:\n          // cv::Mat visimage = visualize_detections(cvimage, result);\n          // cv::imwrite(\"visresult.jpg\", visimage);\n  \n          // -------------------- Model Cloning Demo --------------------\n          auto cloned_detector = detector-\u003eclone();  // Create an independent instance\n          deploy::DetResult clonedresult = cloned_detector-\u003epredict(inputimage);\n  \n          // Verify result consistency\n          std::cout \u003c\u003c cloned_result \u003c\u003c std::endl;\n  \n      } catch (const std::exception\u0026 e) {\n          std::cerr \u003c\u003c \"Program Exception: \" \u003c\u003c e.what() \u003c\u003c std::endl;\n          return EXIT_FAILURE;\n      }\n      return EXIT_SUCCESS;\n  }\n  ```\n\n### 5. Inference Flowchart\n\nBelow is the flowchart of the `predict` method, which illustrates the complete process from input image to output result:\n\n\u003cdiv\u003e\n  \u003cp\u003e\n      \u003cimg width=\"100%\" src=\"./assets/flowsheet.png\"\u003e\u003c/a\u003e\n  \u003c/p\u003e\n\u003c/div\u003e\n\nSimply pass the image to be inferred to the `predict` method. The `predict` method will automatically complete preprocessing, model inference, and post-processing internally, and output the inference results. These results can be further applied to downstream tasks (such as visualization, object tracking, etc.).\n\n\u003e For more deployment examples, please refer to the [Model Deployment Examples](examples) section.\n\n## \u003cdiv align=\"center\"\u003e🖥️ Model Support List\u003c/div\u003e\u003cdiv id=\"support-models\"\u003e\u003c/div\u003e\n\n\u003cdiv align=\"center\"\u003e\n    \u003ctable\u003e\n        \u003ctr\u003e\n            \u003ctd\u003e\n                \u003cimg src='assets/yolo-detect.jpeg' height=\"300\"\u003e\n                \u003ccenter\u003eDetect\u003c/center\u003e\n            \u003c/td\u003e\n            \u003ctd\u003e\n                \u003cimg src='assets/yolo-segment.jpeg' height=\"300\"\u003e\n                \u003ccenter\u003eSegment\u003c/center\u003e\n            \u003c/td\u003e\n        \u003c/tr\u003e\n        \u003ctr\u003e\n            \u003ctd\u003e\n                \u003cimg src='assets/yolo-pose.jpeg' height=\"300\"\u003e\n                \u003ccenter\u003ePose\u003c/center\u003e\n            \u003c/td\u003e\n            \u003ctd\u003e\n                \u003cimg src='assets/yolo-obb.jpeg' height=\"300\"\u003e                                \n                \u003ccenter\u003eOBB\u003c/center\u003e\n            \u003c/td\u003e\n        \u003c/tr\u003e\n    \u003c/table\u003e\n\u003c/div\u003e\n\nSymbol legend: (1)  ✅ : Supported; (2) ❔: In progress; (3) ❎ : Not supported; (4) ❎ : Self-implemented export required for inference. \u003cbr\u003e\n\n\u003cdiv style=\"text-align: center;\"\u003e\n  \u003ctable border=\"1\" style=\"border-collapse: collapse; width: 100%;\"\u003e\n    \u003ctr\u003e\n      \u003cth style=\"text-align: center;\"\u003eTask Scenario\u003c/th\u003e\n      \u003cth style=\"text-align: center;\"\u003eModel\u003c/th\u003e\n      \u003cth style=\"text-align: center;\"\u003eCLI Export\u003c/th\u003e\n      \u003cth style=\"text-align: center;\"\u003eInference Deployment\u003c/th\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003eDetect\u003c/td\u003e\n      \u003ctd\u003e\u003ca href=\"https://github.com/ultralytics/yolov3\"\u003eultralytics/yolov3\u003c/a\u003e\u003c/td\u003e \n      \u003ctd\u003e✅\u003c/td\u003e\n      \u003ctd\u003e✅\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003eDetect\u003c/td\u003e\n      \u003ctd\u003e\u003ca href=\"https://github.com/ultralytics/yolov5\"\u003eultralytics/yolov5\u003c/a\u003e\u003c/td\u003e \n      \u003ctd\u003e✅\u003c/td\u003e\n      \u003ctd\u003e✅\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003eDetect\u003c/td\u003e\n      \u003ctd\u003e\u003ca href=\"https://github.com/meituan/YOLOv6\"\u003emeituan/YOLOv6\u003c/a\u003e\u003c/td\u003e \n      \u003ctd\u003e❎ Refer to \u003ca href=\"https://github.com/meituan/YOLOv6/tree/main/deploy/ONNX#tensorrt-backend-tensorrt-version-800\"\u003eofficial export tutorial\u003c/a\u003e\u003c/td\u003e \n      \u003ctd\u003e✅\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003eDetect\u003c/td\u003e\n      \u003ctd\u003e\u003ca href=\"https://github.com/WongKinYiu/yolov7\"\u003eWongKinYiu/yolov7\u003c/a\u003e\u003c/td\u003e \n      \u003ctd\u003e❎ Refer to \u003ca href=\"https://github.com/WongKinYiu/yolov7#export\"\u003eofficial export tutorial\u003c/a\u003e\u003c/td\u003e \n      \u003ctd\u003e✅\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003eDetect\u003c/td\u003e\n      \u003ctd\u003e\u003ca href=\"https://github.com/WongKinYiu/yolov9\"\u003eWongKinYiu/yolov9\u003c/a\u003e\u003c/td\u003e \n      \u003ctd\u003e❎ Refer to \u003ca href=\"https://github.com/WongKinYiu/yolov9/issues/130#issue-2162045461\"\u003eofficial export tutorial\u003c/a\u003e\u003c/td\u003e \n      \u003ctd\u003e✅\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003eDetect\u003c/td\u003e\n      \u003ctd\u003e\u003ca href=\"https://github.com/THU-MIG/yolov10\"\u003eTHU-MIG/yolov10\u003c/a\u003e\u003c/td\u003e \n      \u003ctd\u003e✅\u003c/td\u003e\n      \u003ctd\u003e✅\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003eDetect\u003c/td\u003e\n      \u003ctd\u003e\u003ca href=\"https://github.com/ultralytics/ultralytics\"\u003eultralytics/ultralytics\u003c/a\u003e\u003c/td\u003e \n      \u003ctd\u003e✅\u003c/td\u003e\n      \u003ctd\u003e✅\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003eDetect\u003c/td\u003e\n      \u003ctd\u003e\u003ca href=\"https://github.com/PaddlePaddle/PaddleDetection\"\u003ePaddleDetection/PP-YOLOE+\u003c/a\u003e\u003c/td\u003e \n      \u003ctd\u003e✅\u003c/td\u003e\n      \u003ctd\u003e✅\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003eSegment\u003c/td\u003e\n      \u003ctd\u003e\u003ca href=\"https://github.com/ultralytics/yolov3\"\u003eultralytics/yolov3\u003c/a\u003e\u003c/td\u003e \n      \u003ctd\u003e✅\u003c/td\u003e\n      \u003ctd\u003e✅\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003eSegment\u003c/td\u003e\n      \u003ctd\u003e\u003ca href=\"https://github.com/ultralytics/yolov5\"\u003eultralytics/yolov5\u003c/a\u003e\u003c/td\u003e \n      \u003ctd\u003e✅\u003c/td\u003e\n      \u003ctd\u003e✅\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003eSegment\u003c/td\u003e\n      \u003ctd\u003e\u003ca href=\"https://github.com/meituan/YOLOv6/tree/yolov6-seg\"\u003emeituan/YOLOv6-seg\u003c/a\u003e\u003c/td\u003e \n      \u003ctd\u003e❎ Implement yourself referring to \u003ca href=\"https://github.com/laugh12321/TensorRT-YOLO/blob/main/tensorrt_yolo/export/head.py\"\u003etensorrt_yolo/export/head.py\u003c/a\u003e\u003c/td\u003e\n      \u003ctd\u003e🟢\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003eSegment\u003c/td\u003e\n      \u003ctd\u003e\u003ca href=\"https://github.com/WongKinYiu/yolov7\"\u003eWongKinYiu/yolov7\u003c/a\u003e\u003c/td\u003e \n      \u003ctd\u003e❎ Implement yourself referring to \u003ca href=\"https://github.com/laugh12321/TensorRT-YOLO/blob/main/tensorrt_yolo/export/head.py\"\u003etensorrt_yolo/export/head.py\u003c/a\u003e\u003c/td\u003e\n      \u003ctd\u003e🟢\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003eSegment\u003c/td\u003e\n      \u003ctd\u003e\u003ca href=\"https://github.com/WongKinYiu/yolov9\"\u003eWongKinYiu/yolov9\u003c/a\u003e\u003c/td\u003e \n      \u003ctd\u003e❎ Implement yourself referring to \u003ca href=\"https://github.com/laugh12321/TensorRT-YOLO/blob/main/tensorrt_yolo/export/head.py\"\u003etensorrt_yolo/export/head.py\u003c/a\u003e\u003c/td\u003e\n      \u003ctd\u003e🟢\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003eSegment\u003c/td\u003e\n      \u003ctd\u003e\u003ca href=\"https://github.com/ultralytics/ultralytics\"\u003eultralytics/ultralytics\u003c/a\u003e\u003c/td\u003e \n      \u003ctd\u003e✅\u003c/td\u003e\n      \u003ctd\u003e✅\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003eClassify\u003c/td\u003e\n      \u003ctd\u003e\u003ca href=\"https://github.com/ultralytics/yolov3\"\u003eultralytics/yolov3\u003c/a\u003e\u003c/td\u003e\n      \u003ctd\u003e✅\u003c/td\u003e\n      \u003ctd\u003e✅\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003eClassify\u003c/td\u003e\n      \u003ctd\u003e\u003ca href=\"https://github.com/ultralytics/yolov5\"\u003eultralytics/yolov5\u003c/a\u003e\u003c/td\u003e\n      \u003ctd\u003e✅\u003c/td\u003e\n      \u003ctd\u003e✅\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003eClassify\u003c/td\u003e\n      \u003ctd\u003e\u003ca href=\"https://github.com/ultralytics/ultralytics\"\u003eultralytics/ultralytics\u003c/a\u003e\u003c/td\u003e\n      \u003ctd\u003e✅\u003c/td\u003e\n      \u003ctd\u003e✅\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003ePose\u003c/td\u003e\n      \u003ctd\u003e\u003ca href=\"https://github.com/ultralytics/ultralytics\"\u003eultralytics/ultralytics\u003c/a\u003e\u003c/td\u003e\n      \u003ctd\u003e✅\u003c/td\u003e\n      \u003ctd\u003e✅\u003c/td\u003e\n    \u003c/tr\u003e\n    \u003ctr\u003e\n      \u003ctd\u003eOBB\u003c/td\u003e\n      \u003ctd\u003e\u003ca href=\"https://github.com/ultralytics/ultralytics\"\u003eultralytics/ultralytics\u003c/a\u003e\u003c/td\u003e\n      \u003ctd\u003e✅\u003c/td\u003e\n      \u003ctd\u003e✅\u003c/td\u003e\n    \u003c/tr\u003e\n  \u003c/table\u003e\n\u003c/div\u003e\n\n\u003ch2 align=\"center\"\u003e🌟 Sponsorship \u0026 Support\u003c/h2\u003e\n\nOpen-source projects thrive on support. If this project has been helpful to you, consider sponsoring the author. Your support is the greatest motivation for continued development!\n\n\u003cdiv align=\"center\"\u003e\n  \u003ca href=\"https://afdian.com/a/laugh12321\"\u003e\n    \u003cimg width=\"200\" src=\"https://pic1.afdiancdn.com/static/img/welcome/button-sponsorme.png\" alt=\"Sponsor Me\"\u003e\n  \u003c/a\u003e\n\u003c/div\u003e\n\n---\n\n🙏 **A Heartfelt Thank You to Our Supporters and Sponsors**:\n\n\u003e [!NOTE]\n\u003e\n\u003e The following is a list of sponsors automatically generated by GitHub Actions, updated daily ✨.\n\n\u003cdiv align=\"center\"\u003e\n  \u003ca target=\"_blank\" href=\"https://afdian.com/a/laugh12321\"\u003e\n    \u003cimg alt=\"Sponsors List\" src=\"https://github.com/laugh12321/sponsor/blob/main/sponsors.svg?raw=true\"\u003e\n  \u003c/a\u003e\n\u003c/div\u003e\n\n## \u003cdiv align=\"center\"\u003e📄 License\u003c/div\u003e\n\nTensorRT-YOLO is licensed under the **GPL-3.0 License**, an [OSI-approved](https://opensource.org/licenses/) open-source license that is ideal for students and enthusiasts, fostering open collaboration and knowledge sharing. Please refer to the [LICENSE](https://github.com/laugh12321/TensorRT-YOLO/blob/master/LICENSE) file for more details.\n\nThank you for choosing TensorRT-YOLO; we encourage open collaboration and knowledge sharing, and we hope you comply with the relevant provisions of the open-source license.\n\n## \u003cdiv align=\"center\"\u003e📞 Contact\u003c/div\u003e\n\nFor bug reports and feature requests regarding TensorRT-YOLO, please visit [GitHub Issues](https://github.com/laugh12321/TensorRT-YOLO/issues)!\n\n## \u003cdiv align=\"center\"\u003e🙏 Thanks\u003c/div\u003e\n\n\u003cdiv align=\"center\"\u003e\n\u003ca href=\"https://hellogithub.com/repository/942570b550824b1b9397e4291da3d17c\" target=\"_blank\"\u003e\u003cimg src=\"https://api.hellogithub.com/v1/widgets/recommend.svg?rid=942570b550824b1b9397e4291da3d17c\u0026claim_uid=2AGzE4dsO8ZUD9R\u0026theme=neutral\" alt=\"Featured｜HelloGitHub\" style=\"width: 250px; height: 54px;\" width=\"250\" height=\"54\" /\u003e\u003c/a\u003e\n\u003c/div\u003e\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Flaugh12321%2FTensorRT-YOLO","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Flaugh12321%2FTensorRT-YOLO","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Flaugh12321%2FTensorRT-YOLO/lists"}