๐Ÿš€ Easier & Faster YOLO Deployment Toolkit for NVIDIA ๐Ÿ› ๏ธ

1.4K stars 149 forks 1.4K watchers C++ GNU General Public License v3.0
computer-vision image-classification instance-segmentation object-detection pose-estimation rotated-object-detection tensorrt tensorrt-inference tensorrt10 ultralytics yolo yolo11 yolo12 yolov10 yolov3 yolov5 yolov6 yolov7 yolov8 yolov9
5 Open Issues Need Help Last updated: Sep 13, 2025

Open Issues Need Help

View All on GitHub

AI Summary: The user is encountering a C4819 warning during the compilation of a `warpaffine.cu` file within a TensorRT-YOLO project. This warning indicates a character encoding issue, suggesting the file contains characters not representable in the current code page and should be saved in Unicode format.

Complexity: 2/5
help wanted

๐Ÿš€ Easier & Faster YOLO Deployment Toolkit for NVIDIA ๐Ÿ› ๏ธ

C++
#computer-vision#image-classification#instance-segmentation#object-detection#pose-estimation#rotated-object-detection#tensorrt#tensorrt-inference#tensorrt10#ultralytics#yolo#yolo11#yolo12#yolov10#yolov3#yolov5#yolov6#yolov7#yolov8#yolov9

AI Summary: A user is encountering a 'Segmentation fault (core dumped)' error when attempting to perform inference using TensorRT 10, whether through Python or the command-line interface. They are seeking guidance on how to identify the specific cause of the error.

Complexity: 4/5
help wanted

๐Ÿš€ Easier & Faster YOLO Deployment Toolkit for NVIDIA ๐Ÿ› ๏ธ

C++
#computer-vision#image-classification#instance-segmentation#object-detection#pose-estimation#rotated-object-detection#tensorrt#tensorrt-inference#tensorrt10#ultralytics#yolo#yolo11#yolo12#yolov10#yolov3#yolov5#yolov6#yolov7#yolov8#yolov9

AI Summary: The task is to determine if the existing TensorRT-YOLO project allows for dynamic adjustment of IOU and confidence thresholds during inference, rather than only during model export. If not currently supported, the feasibility of adding this feature needs to be assessed.

Complexity: 4/5
duplicate help wanted

๐Ÿš€ Easier & Faster YOLO Deployment Toolkit for NVIDIA ๐Ÿ› ๏ธ

C++
#computer-vision#image-classification#instance-segmentation#object-detection#pose-estimation#rotated-object-detection#tensorrt#tensorrt-inference#tensorrt10#ultralytics#yolo#yolo11#yolo12#yolov10#yolov3#yolov5#yolov6#yolov7#yolov8#yolov9

AI Summary: The task is to troubleshoot a Python ImportError related to the 'TypeAlias' name not being found in the 'typing' module, specifically within the context of the TensorRT-YOLO project. The user is using Python 3.9 and wants to know if the error can be resolved without upgrading to Python 3.10, given that they've already updated the TensorRT-YOLO version.

Complexity: 4/5
help wanted

๐Ÿš€ Easier & Faster YOLO Deployment Toolkit for NVIDIA ๐Ÿ› ๏ธ

C++
#computer-vision#image-classification#instance-segmentation#object-detection#pose-estimation#rotated-object-detection#tensorrt#tensorrt-inference#tensorrt10#ultralytics#yolo#yolo11#yolo12#yolov10#yolov3#yolov5#yolov6#yolov7#yolov8#yolov9

AI Summary: Determine if the TensorRT-YOLO C++ codebase is compatible with the C++17 standard and whether it's incompatible with C++14. This involves reviewing the code for any C++17-specific features or constructs that might cause compilation errors under C++14.

Complexity: 3/5
help wanted

๐Ÿš€ Easier & Faster YOLO Deployment Toolkit for NVIDIA ๐Ÿ› ๏ธ

C++
#computer-vision#image-classification#instance-segmentation#object-detection#pose-estimation#rotated-object-detection#tensorrt#tensorrt-inference#tensorrt10#ultralytics#yolo#yolo11#yolo12#yolov10#yolov3#yolov5#yolov6#yolov7#yolov8#yolov9