使用 DeepStream SDK 和 TensorRT 在 NVIDIA Jetson 上运行 Ultralytics YOLO26
观看: 如何在 Jetson Orin NX 上通过 NVIDIA Deepstream 使用 Ultralytics YOLO26 模型 🚀
本指南将介绍如何使用 DeepStream SDK 和 TensorRT 在 NVIDIA Jetson 设备上部署 Ultralytics YOLO26。我们在此使用 TensorRT,以最大限度地提升 Jetson 平台上的推理性能。
本指南将带你了解 YOLO26 的 DeepStream 配置、INT8 校准、多流设置和基准测试结果。

本指南已在运行 JetPack JP6.1 版本的 NVIDIA Jetson Orin Nano Super Developer Kit、基于 NVIDIA Jetson Orin NX 16GB 且运行 JetPack JP5.1.3 版本的 Seeed Studio reComputer J4012,以及基于 NVIDIA Jetson Nano 4GB 且运行 JetPack JP4.6.4 版本的 Seeed Studio reComputer J1020 v2 上进行测试。本指南预计适用于整个 NVIDIA Jetson 硬件系列,包括最新和旧款设备。
什么是 NVIDIA DeepStream?#
NVIDIA DeepStream SDK 是一套基于 GStreamer 的完整流式分析工具包,用于基于 AI 的多传感器处理、视频、音频和图像理解。它非常适合构建 IVA(智能视频分析)应用和服务的视觉 AI 开发者、软件合作伙伴、初创公司和 OEM。你可以创建流处理流水线,将神经网络与跟踪、视频编码/解码和视频渲染等其他复杂处理任务结合起来。这些流水线能够对视频、图像和传感器数据进行实时分析。DeepStream 的多平台支持让你能够更快速、更轻松地在本地、边缘端和云端开发视觉 AI 应用和服务。
前置条件#
开始阅读本指南之前:
- 请访问我们的文档 快速入门指南:使用 Ultralytics YOLO26 的 NVIDIA Jetson,为你的 NVIDIA Jetson 设备配置 Ultralytics YOLO26
- 根据 JetPack 版本安装 DeepStream SDK
- 对于 JetPack 4.6.4,请安装 DeepStream 6.0.1
- 对于 JetPack 5.1.3,请安装 DeepStream 6.3
- 对于 JetPack 6.1,请安装 DeepStream 7.1
- 对于 JetPack 7.1,请安装 DeepStream 9.0
本指南采用 Debian 软件包方式在 Jetson 设备上安装 DeepStream SDK。你也可以访问 Jetson 上的 DeepStream SDK(存档),获取 DeepStream 的旧版本。
YOLO26 的 DeepStream 配置#
这里使用了 marcoslucianops/DeepStream-Yolo GitHub 仓库,该仓库为 YOLO 模型提供 NVIDIA DeepStream SDK 支持。感谢 marcoslucianops 作出的贡献。
-
安装 Ultralytics 及必要的依赖项
pip install ultralytics onnx onnxslim -
克隆 DeepStream-Yolo 仓库
cd ~ git clone https://github.com/marcoslucianops/DeepStream-Yolo -
从 YOLO26 项目下载你选择的 Ultralytics YOLO26 检测模型(.pt)。这里我们使用 yolo26s.pt。
wget https://github.com/ultralytics/assets/releases/download/v8.4.0/yolo26s.pt
你也可以使用自定义训练的 YOLO26 模型。
-
将模型转换为 ONNX,并写入
labels.txt文件,DeepStream 会从中读取类别名称from ultralytics import YOLO model = YOLO("yolo26s.pt") model.export(format="onnx", nms=False, agnostic_nms=True) # 创建了 'yolo26s.onnx' with open("labels.txt", "w") as f: f.write("\n".join(model.names.values()))
nms=False 会导出无 NMS 的检测头,而 agnostic_nms=True 会让每个检测结果只保留一个类别,这样 DeepStream 配置中的 cluster-mode=4(无聚类)就不会重复绘制边界框。添加 imgsz=1280 可更改推理尺寸(默认值:640);添加 batch=4 可将批次大小设为 4(导出的批次大小是静态的,因此必须与 DeepStream 配置中的 batch-size 匹配);添加 opset=12 可用于 TensorRT 8.2 或更早版本(DeepStream 6.0.1 及更早版本)。完整列表请参阅导出参数。
-
将生成的
.onnx模型文件和labels.txt文件复制到DeepStream-Yolo文件夹cp yolo26s.onnx labels.txt ~/DeepStream-Yolo cd ~/DeepStream-Yolo -
根据已安装的 JetPack 版本设置 CUDA 版本
对于 JetPack 4.6.4:
export CUDA_VER=10.2对于 JetPack 5.1.3:
export CUDA_VER=11.4对于 JetPack 6.1:
export CUDA_VER=12.6对于 JetPack 7.1:
export CUDA_VER=13.0 -
编译库
make -C nvdsinfer_custom_impl_Yolo clean && make -C nvdsinfer_custom_impl_Yolo -
根据你的模型编辑
config_infer_primary_yolo26.txt文件(适用于包含 80 个类别的 YOLO26s)[property] ... onnx-file=yolo26s.onnx ... num-detected-classes=80 ... parse-bbox-func-name=NvDsInferParseYolo ...
YOLO26 会对输入进行缩放并在中心填充,且不使用 NMS。为获得最佳精度,请将以下内容添加到 config_infer_primary_yolo26.txt 的 [property] 部分:
[property]
...
maintain-aspect-ratio=1
symmetric-padding=1
cluster-mode=4
...-
编辑
deepstream_app_config文件... [primary-gie] ... config-file=config_infer_primary_yolo26.txt -
你也可以在
deepstream_app_config文件中更改视频源。这里加载的是默认视频文件... [source0] ... uri=file:///opt/nvidia/deepstream/deepstream/samples/streams/sample_1080p_h264.mp4
运行推理#
deepstream-app -c deepstream_app_config.txt生成 TensorRT 引擎文件可能需要很长时间,之后才能开始推理。

要将模型转换为 FP16 精度,请在 config_infer_primary_yolo26.txt 中设置 model-engine-file=model_b1_gpu0_fp16.engine 和 network-mode=2
INT8 校准#
要使用 INT8 精度进行推理,请按照以下步骤操作:
目前 INT8 无法与 TensorRT 10.x 配合使用。本指南的这一部分已使用 TensorRT 8.x 进行测试,预计可正常运行。
-
设置
OPENCV环境变量export OPENCV=1 -
编译库
make -C nvdsinfer_custom_impl_Yolo clean && make -C nvdsinfer_custom_impl_Yolo -
对于 COCO 数据集,请下载 val2017,解压后移动到
DeepStream-Yolo文件夹 -
新建一个目录,用于存放校准图像
mkdir calibration -
运行以下命令,从 COCO 数据集中随机选取 1000 张图像用于校准
for jpg in $(ls -1 val2017/*.jpg | sort -R | head -1000); do cp ${jpg} calibration/ done
NVIDIA 建议至少使用 500 张图像以获得良好精度。本例选择 1000 张图像以获得更高精度(通常图像越多,精度越高)。你可以通过 head -1000 设置数量。例如,要使用 2000 张图像,请设置 head -2000。此过程可能需要很长时间。
-
创建
calibration.txt文件,并将所有选中的图像列入其中realpath calibration/*jpg > calibration.txt -
设置环境变量
export INT8_CALIB_IMG_PATH=calibration.txt export INT8_CALIB_BATCH_SIZE=1
较高的 INT8_CALIB_BATCH_SIZE 值可带来更高精度和更快的校准速度。请根据你的 GPU 内存进行设置。
-
更新
config_infer_primary_yolo26.txt文件从
... model-engine-file=model_b1_gpu0_fp32.engine #int8-calib-file=calib.table ... network-mode=0 ...到
... model-engine-file=model_b1_gpu0_int8.engine int8-calib-file=calib.table ... network-mode=1 ...
运行 INT8 推理#
运行相同的命令来构建 INT8 引擎并开始推理:
deepstream-app -c deepstream_app_config.txt多流设置#
观看: 如何在 Jetson Orin 上使用 NVIDIA DeepStream 和 Ultralytics YOLO26 执行多路流推理 🚀
要在单个 DeepStream 应用程序中设置多个流,请对 deepstream_app_config.txt 文件进行以下更改:
-
根据要使用的流数量,更改行数和列数以构建网格显示。例如,对于 4 个流,可以添加 2 行和 2 列。
[tiled-display] rows=2 columns=2 -
为每个流添加单独的
[sourceN]组,每组都要有自己的uri和num-sources=1。[source0] enable=1 type=3 uri=file:///path/to/video1.mp4 num-sources=1 [source1] enable=1 type=3 uri=file:///path/to/video2.mp4 num-sources=1 [source2] enable=1 type=3 uri=file:///path/to/video3.mp4 num-sources=1 [source3] enable=1 type=3 uri=file:///path/to/video4.mp4 num-sources=1
运行多流推理#
运行相同的命令,在平铺显示中启动所有流:
deepstream-app -c deepstream_app_config.txt
基准测试结果#
以下基准测试总结了 YOLO11 模型在 NVIDIA Jetson Orin NX 16GB 上使用不同 TensorRT 精度级别、输入尺寸为 640x640 时的性能。YOLO26 使用上文介绍的相同 DeepStream 导出和推理工作流。
对比图表#

详细对比表#
| 格式 | 状态 | 推理时间 (ms/im) |
|---|---|---|
| TensorRT (FP32) | ✅ | 8.64 |
| TensorRT (FP16) | ✅ | 5.27 |
| TensorRT (INT8) | ✅ | 4.54 |
致谢#
本指南最初由 Seeed Studio 的朋友 Lakshantha 和 Elaine 编写。
常见问题#
要在 NVIDIA Jetson 设备上设置 Ultralytics YOLO26,你需要先安装与你的 JetPack 版本兼容的 DeepStream SDK。请按照我们的快速入门指南中的分步说明,为 YOLO26 部署配置 NVIDIA Jetson。
使用 TensorRT 与 YOLO26 配合,可以优化模型以进行推理,显著降低延迟并提高 NVIDIA Jetson 设备上的吞吐量。TensorRT 通过层融合、精度校准和内核自动调优,提供高性能、低延迟的深度学习推理。这能带来更快速、更高效的执行,尤其适用于视频分析和自主机器等实时应用。
可以,使用 DeepStream SDK 和 TensorRT 部署 Ultralytics YOLO26 的指南兼容整个 NVIDIA Jetson 系列。这包括搭载 JetPack 5.1.3 的 Jetson Orin NX 16GB,以及搭载 JetPack 4.6.4 的 Jetson Nano 4GB。详细步骤请参阅 YOLO26 的 DeepStream 配置部分。
使用 Ultralytics 导出器导出带有无 NMS 检测头的模型,并写入
labels.txt文件,DeepStream 会从中读取类别名称:from ultralytics import YOLO model = YOLO("yolo26s.pt") model.export(format="onnx", nms=False, agnostic_nms=True) # 创建了 'yolo26s.onnx' with open("labels.txt", "w") as f: f.write("\n".join(model.names.values()))对于 TensorRT 8.2 或更早版本(DeepStream 6.0.1 及更早版本),请添加
opset=12。有关模型转换的更多详情,请参阅我们的模型导出部分。要运行 INT8 推理,请使用具有代表性的图像集校准模型,并将 DeepStream 配置切换为 INT8 模式。下载 COCO val2017 图像,选取约 1000 张校准图像,设置
INT8_CALIB_IMG_PATH和INT8_CALIB_BATCH_SIZE环境变量,然后使用model-engine-file=model_b1_gpu0_int8.engine、int8-calib-file=calib.table和network-mode=1更新config_infer_primary_yolo26.txt。完整步骤请参阅 INT8 校准部分。目前,INT8 需要 TensorRT 8.x。要在单个 DeepStream 应用程序中处理多个流,请编辑
deepstream_app_config.txt文件,添加平铺显示网格并列出每个源 URI。在[tiled-display]下设置rows和columns以构建网格;为每个流添加单独的[sourceN]组,并为其设置各自的uri和num-sources=1;然后调整网格以适应流的数量。完整示例请参阅多流设置部分。YOLO11 模型在 NVIDIA Jetson Orin NX 16GB 上的性能因 TensorRT 精度级别而异。例如,YOLO11s 模型可达到:
- FP32 精度:14.53 ms/im,68.8 FPS
- FP16 精度:7.91 ms/im,126 FPS
- INT8 精度:6.05 ms/im,165 FPS
这些结果展示了 FP16 和 INT8 TensorRT 引擎能在 NVIDIA Jetson 硬件上将 YOLO11 模型的运行速度提升多少。
评论145
提出问题、分享结果,或帮助其他人。
Llinz24
hi!I encountered the following issue when using NVIDIA's official DeepStream 8.0 gc-triton-devel image to implement deepstream-yolo: I cannot compile the library, and it prompts that my Tensor version is too new. Is there any solution? Could it be that the current code hasn't fully adapted to DeepStream 8.0? ( root@LGDX:/workspace/DeepStream-Yolo# export CUDA_VER=12.8 root@LGDX:/workspace/DeepStream-Yolo# make -C nvdsinfer_custom_impl_Yolo clean && make -C nvdsinfer_custom_impl_Yolo make: Entering directory '/workspace/DeepStream-Yolo/nvdsinfer_custom_impl_Yolo' rm -rf libnvdsinfer_custom_impl_Yolo.so rm -rf nvdsinfer_yolo_engine.o nvdsparsebbox_Yolo.o utils.o yolo.o yoloPlugins.o layers/activation_layer.o layers/batchnorm_layer.o layers/channels_layer.o layers/convolutional_layer.o layers/deconvolutional_layer.o layers/implicit_layer.o layers/pooling_layer.o layers/reorg_layer.o layers/route_layer.o layers/sam_layer.o layers/shortcut_layer.o layers/slice_layer.o layers/upsample_layer.o nvdsparsebbox_Yolo_cuda.o yoloForward.o yoloForward_nc.o yoloForward_v2.o make: Leaving directory '/workspace/DeepStream-Yolo/nvdsinfer_custom_impl_Yolo' make: Entering directory '/workspace/DeepStream-Yolo/nvdsinfer_custom_impl_Yolo' g++ -c -o nvdsinfer_yolo_engine.o -Wall -std=c++11 -shared -fPIC -Wno-error=deprecated-declarations -I/opt/nvidia/deepstream/deepstream/sources/includes -I/usr/local/cuda-12.8/include nvdsinfer_yolo_engine.cpp g++ -c -o nvdsparsebbox_Yolo.o -Wall -std=c++11 -shared -fPIC -Wno-error=deprecated-declarations -I/opt/nvidia/deepstream/deepstream/sources/includes -I/usr/local/cuda-12.8/include nvdsparsebbox_Yolo.cpp In file included from nvdsparsebbox_Yolo.cpp:26: nvdsparsebbox_Yolo.cpp: In function 'void checkFunc_NvDsInferParseYolo(NvDsInferParseCustomFunc)': /opt/nvidia/deepstream/deepstream/sources/includes/nvdsinfer_custom_impl.h:221:17: warning: infinite recursion detected [-Winfinite-recursion] 221 | static void checkFunc_ ## customParseFunc (NvDsInferParseCustomFunc func = customParseFunc)
| ^~~~~~~~~~ nvdsparsebbox_Yolo.cpp:129:1: note: in expansion of macro 'CHECK_CUSTOM_PARSE_FUNC_PROTOTYPE' 129 | CHECK_CUSTOM_PARSE_FUNC_PROTOTYPE(NvDsInferParseYolo); | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ /opt/nvidia/deepstream/deepstream/sources/includes/nvdsinfer_custom_impl.h:222:41: note: recursive call 222 | { checkFunc_ ## customParseFunc (); };
| ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~ nvdsparsebbox_Yolo.cpp:129:1: note: in expansion of macro 'CHECK_CUSTOM_PARSE_FUNC_PROTOTYPE' 129 | CHECK_CUSTOM_PARSE_FUNC_PROTOTYPE(NvDsInferParseYolo); | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ g++ -c -o utils.o -Wall -std=c++11 -shared -fPIC -Wno-error=deprecated-declarations -I/opt/nvidia/deepstream/deepstream/sources/includes -I/usr/local/cuda-12.8/include utils.cpp g++ -c -o yolo.o -Wall -std=c++11 -shared -fPIC -Wno-error=deprecated-declarations -I/opt/nvidia/deepstream/deepstream/sources/includes -I/usr/local/cuda-12.8/include yolo.cpp In file included from yolo.cpp:29: yoloPlugins.h:53:59: warning: 'IPluginV2DynamicExt' is deprecated [-Wdeprecated-declarations] 53 | nvinfer1::IPluginV2DynamicExt* clone() const noexcept override; | ^~~~~~~~ In file included from /usr/include/x86_64-linux-gnu/NvInfer.h:22, from /usr/include/x86_64-linux-gnu/NvOnnxParser.h:21, from yolo.cpp:26: /usr/include/x86_64-linux-gnu/NvInferRuntime.h:406:22: note: declared here 406 | class TRT_DEPRECATED IPluginV2DynamicExt : public nvinfer1::IPluginV2Ext | ^~~~~~~~~~~~~~~~~~~ yoloPlugins.h:124:9: warning: 'IPluginV2DynamicExt' is deprecated [-Wdeprecated-declarations] 124 | override { | ^~~~~~~~ /usr/include/x86_64-linux-gnu/NvInferRuntime.h:406:22: note: declared here 406 | class TRT_DEPRECATED IPluginV2DynamicExt : public nvinfer1::IPluginV2Ext | ^~~~~~~~~~~~~~~~~~~ yoloPlugins.h:130:18: warning: 'IPluginV2DynamicExt' is deprecated [-Wdeprecated-declarations] 130 | noexcept override { | ^~~~~~~~ /usr/include/x86_64-linux-gnu/NvInferRuntime.h:406:22: note: declared here 406 | class TRT_DEPRECATED IPluginV2DynamicExt : public nvinfer1::IPluginV2Ext | ^~~~~~~~~~~~~~~~~~~ yolo.cpp: In member function 'nvinfer1::ICudaEngine* Yolo::createEngine(nvinfer1::IBuilder*, nvinfer1::IBuilderConfig*)': yolo.cpp:71:76: error: 'kEXPLICIT_BATCH' is not a member of 'nvinfer1::NetworkDefinitionCreationFlag' 71 | tic_cast<uint32_t>(nvinfer1::NetworkDefinitionCreationFlag::kEXPLICIT_BATCH); | ^~~~~~~~~~~~~~~In file included from /usr/include/c++/13/cassert:44, from layers/../utils.h:32, from layers/route_layer.h:9, from yolo.h:39, from yolo.cpp:28: yolo.cpp:155:21: error: 'class nvinfer1::IBuilder' has no member named 'platformHasFastFp16' 155 | assert(builder->platformHasFastFp16()); | ^~~~~~~~~~~~~~~~~~~ yolo.cpp:156:44: error: 'kFP16' is not a member of 'nvinfer1::BuilderFlag' 156 | config->setFlag(nvinfer1::BuilderFlag::kFP16); | ^~~~~ yolo.cpp:159:21: error: 'class nvinfer1::IBuilder' has no member named 'platformHasFastInt8' 159 | assert(builder->platformHasFastInt8()); | ^~~~~~~~~~~~~~~~~~~ yolo.cpp:160:44: error: 'kINT8' is not a member of 'nvinfer1::BuilderFlag' 160 | config->setFlag(nvinfer1::BuilderFlag::kINT8); | ^~~~~ yolo.cpp: In member function 'NvDsInferStatus Yolo::buildYoloNetwork(std::vector&, nvinfer1::INetworkDefinition&)': yolo.cpp:481:36: warning: 'IPluginV2DynamicExt' is deprecated [-Wdeprecated-declarations] 481 | nvinfer1::IPluginV2DynamicExt* yoloPlugin = new YoloLayer(m_InputW, m_InputH, m_NumClasses, m_NewCoords, | ^~~~~~~~~~ /usr/include/x86_64-linux-gnu/NvInferRuntime.h:406:22: note: declared here 406 | class TRT_DEPRECATED IPluginV2DynamicExt : public nvinfer1::IPluginV2Ext | ^~~~~~~~~~~~~~~~~~~ yolo.cpp:484:31: warning: 'IPluginV2Layer' is deprecated [-Wdeprecated-declarations] 484 | nvinfer1::IPluginV2Layer* yolo = network.addPluginV2(yoloTensorInputs, m_YoloCount, yoloPlugin); | ^~~~ /usr/include/x86_64-linux-gnu/NvInfer.h:2542:22: note: declared here 2542 | class TRT_DEPRECATED IPluginV2Layer : public ILayer | ^~~~~~~~~~~~~~ yolo.cpp:484:57: warning: 'nvinfer1::IPluginV2Layer nvinfer1::INetworkDefinition::addPluginV2(nvinfer1::ITensor* const*, int32_t, nvinfer1::IPluginV2&)' is deprecated [-Wdeprecated-declarations] 484 | nvinfer1::IPluginV2Layer* yolo = network.addPluginV2(yoloTensorInputs, m_YoloCount, yoloPlugin); | ~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ /usr/include/x86_64-linux-gnu/NvInfer.h:8859:36: note: declared here 8859 | TRT_DEPRECATED IPluginV2Layer addPluginV2(ITensor* const* inputs, int32_t nbInputs, IPluginV2& plugin) noexcept | ^~~~~~~~~~~ make: *** [Makefile:86: yolo.o] Error 1 make: Leaving directory '/workspace/DeepStream-Yolo/nvdsinfer_custom_impl_Yolo' root@LGDX:/workspace/DeepStream-Yolo#)
Glenn JocherUltralyticsYes, those build errors are from a TensorRT API mismatch, not your commands: the
DeepStream-Yolopath used in this doc was tested with the DeepStream/JetPack combinations listed in the Jetson + DeepStream guide, andDeepStream 8.0 gc-triton-develis using newer TensorRT APIs where symbols likekEXPLICIT_BATCH,kFP16, andplatformHasFastFp16()are gone, so it will not compile unchanged; for now please use a supported version combo from the guide or a direct TensorRT export workflow, and if you want to stay on DS 8.0 please share the exactDeepStream-Yolocommit plus your TensorRT version here so we can point out the required upstream patch areas.Llinz24
root@LGDX:/workspace/DeepStream-Yolo# dpkg -l | grep tensorrt ii tensorrt-dev 11.0.0.114-1+cuda13.2 amd64 Meta package for TensorRT development libraries
If I plan to use yolo11-pose and yolo26, what kind of adjustments do I need?
Glenn JocherUltralytics(已编辑)For
YOLO11-pose, use DeepStream-Yolo-Pose withconfig_infer_primary_yolo11_pose.txt; forYOLO26detection, use the main DeepStream-Yolo repo and export through its updatedexport_yolo26.pypath or export with thenmsargument described in the End-to-End Detection guide for the classic DeepStream parser flow. The bigger blocker is your environment: DeepStream 8.0 on x86 is validated with TensorRT 10.9.0.34, while your container has TensorRT11.0.0, and TRT 11 removed the older builder-flag/calibrator APIs, so those external DeepStream custom libs need a TRT11 migration before they will compile; the quickest path is a TRT 10.9 DeepStream 8.0 stack plus the latest repo files, and the relevant Ultralytics export compatibility notes are in Export Settings.
UserWhy can't I see my class name in the block boxes with multiple RTSP video streams (like in the single-stream example)?
Is there a way to fix this?
Glenn JocherUltralyticsYes—if you’re using
deepstream-appin tiled multistream mode, bbox labels are often hidden unless you start it with-t, so trydeepstream-app -c deepstream_app_config.txt -t, then confirm[osd]hasdisplay-text=1and increasetext-sizeif the tiles are small; please keep Jetson/DeepStream questions here in Discussions so the team can track them, and the Jetson + DeepStream guide is the right reference alongside NVIDIA’s DeepStream reference app docs. (forums.developer.nvidia.com)
Llinz24
Is YOLO26-seg currently supported for direct deployment and execution in DeepStream? Are there any available tutorials or reference solutions?
Glenn JocherUltralyticsNot in the published direct DeepStream flow yet—the current Jetson + DeepStream guide is detection-focused and uses
DeepStream-Yolo, while the available segmentation reference is DeepStream-Yolo-Seg, which currently listsYOLO11-SegandYOLOv8-Segsupport rather thanYOLO26-seg, so there isn’t an official YOLO26-seg DeepStream tutorial at the moment; feel free to keep the details here in Discussions and we can help with the closest path forward. (docs.ultralytics.com)Llinz24
I'm running into an issue with YOLO11-seg + DeepStream and could use some help 🙏 So I exported my YOLO11-seg model to ONNX and then to a TensorRT engine. When I run deepstream-app, it does find and load the engine just fine — but the segmentation part just doesn't work. I'm not seeing any masks or even bounding boxes on the video output. Any idea what might be going wrong here? Also, when I use the deepstream-yolo export script to get the ONNX file, the output only shows one node: 1 OUTPUT kFLOAT output 100x102406 But from what I've seen with official YOLO-seg exports (like from Ultralytics), there should usually be two outputs — one for boxes/classes and another for the masks. That mismatch feels like it might be part of the problem? Has anyone run into this before? Any tips on: How to export the ONNX correctly for YOLO-seg (maybe a flag I'm missing?) What to tweak in the DeepStream config (config_infer_primary.txt, parser settings, etc.) to make it actually show the segmentation results? Thanks in advance!
Glenn JocherUltralyticsYes, a single-output ONNX is very likely the blocker here: for YOLO11-seg, DeepStream needs the segmentation export with both detection and mask-proto outputs, so a detection-style ONNX/parser will often load the engine but show neither boxes nor masks; please re-export from the Ultralytics side using the workflow in the TensorRT export docs, make sure you are using the segmentation-specific DeepStream parser/config rather than the detection one, and if it still fails please paste the full
deepstream-applog plusyolo checksoutput here so we can verify the tensor names, parser, and engine match.
aneesarom
Below is the result from deepstream
**PERF: FPS 0 (Avg) **PERF: 53.80 (58.72) **PERF: 60.00 (59.38) **PERF: 60.00 (59.59) **PERF: 60.00 (59.69)
But i can get around 85 fps using ultralytics inference. why is that so. what is the need of using deepstream then. or i am doing anything wrong?
Glenn JocherUltralyticsThose
PERFnumbers fromdeepstream-appare end-to-end pipeline FPS and are often capped by your input stream rate and/or display sync (so a steady ~60 usually means you’re running “real-time/render-limited,” not model-limited), while your Ultralytics run is likely closer to “pure inference throughput”; DeepStream’s value is when you need a full GStreamer pipeline (decode/scale/batching/multi-stream/OSD/encode/RTSP), so for a fair comparison try a headless config (disable OSD/tiler and use fakesink withsync=0) as described in the Ultralytics YOLO26 DeepStream on Jetson guide and make sure you’re matching TensorRT precision/imgszlike in the TensorRT export docs:[osd] enable=0 [tiled-display] enable=0 [sink0] type=1 sync=0If you paste your
deepstream_app_config.txt[source0]+[sink0]+[primary-gie]sections (and JetPack/DeepStream versions) here, we can point out what’s throttling it.1aneesarom
@glenn-jocher Thank you so much for your reply. Now i understood and can see 120 Fps with this setting. But i have a doubt when do we need deepstream over normal ultralytics inferencing. Can you explain it
Glenn JocherUltralyticsDeepStream is worth using when you need a production GStreamer pipeline around the model—hardware-accelerated decode/encode, zero-copy GPU memory, batching across multiple cameras/streams, RTSP ingest/egress, OSD/tiling, tracking, and easy composition with other DeepStream plugins—whereas “plain” Ultralytics inference is often the simplest/fastest path for single-stream or custom Python loops that just need model outputs; the DeepStream on Jetson guide and the TensorRT export docs are the best references, and feel free to drop your specific target pipeline (sources/sinks/stream count/latency needs) here so we can sanity-check the recommended setup.
Saarulog
Hello,
Thank you for all your help to my issues.
I have a usb camera with the below format and fails to run. How do I get deepstream to accept or convert the format below to the needed format for display.
Deepstream 7.1/Jetpack 6.2.1
deepstream app text file:
[soruce0] camera- camera-fps-n=9 camera-fps-d=1
v4l2-ctl -d /dev/video0 --list-formats-ext
ioctl: VIDIOC_ENUM_FMT Type: Video Capture
[0]: 'UYVY' (UYVY 4:2:2) Size: Discrete 80x60 Interval: Discrete 0.111s (9.000 fps) [1]: 'Y16 ' (16-bit Greyscale) Size: Discrete 80x60 Interval: Discrete 0.111s (9.000 fps) Size: Discrete 80x63 Interval: Discrete 0.111s (9.000 fps) [2]: 'GREY' (8-bit Greyscale) Size: Discrete 80x60 Interval: Discrete 0.111s (9.000 fps) [3]: 'RGBP' (16-bit RGB 5-6-5) Size: Discrete 80x60 Interval: Discrete 0.111s (9.000 fps) [4]: 'BGR3' (24-bit BGR 8-8-8) Size: Discrete 80x60 Interval: Discrete 0.111s (9.000 fps)GST_DEBUG=3 deepstream-app -c deepstream_app_config_yolo26_1usbcam.txt
Setting min object dimensions as 16x16 instead of 1x1 to support VIC compute mode. WARNING: [TRT]: Using an engine plan file across different models of devices is not recommended and is likely to affect performance or even cause errors. 0:00:00.401328266 6238 0xaaaad1ab46c0 INFO nvinfer gstnvinfer.cpp:684:gst_nvinfer_logger:<primary_gie> NvDsInferContext[UID 1]: Info from NvDsInferContextImpl::deserializeEngineAndBackend() <nvdsinfer_context_impl.cpp:2092> [UID = 1]: deserialized trt engine from :/home/shamsee/DeepStream-Yolo/model_b1_gpu0_fp32.engine Implicit layer support has been deprecated INFO: [Implicit Engine Info]: layers num: 0
0:00:00.401518311 6238 0xaaaad1ab46c0 INFO nvinfer gstnvinfer.cpp:684:gst_nvinfer_logger:<primary_gie> NvDsInferContext[UID 1]: Info from NvDsInferContextImpl::generateBackendContext() <nvdsinfer_context_impl.cpp:2195> [UID = 1]: Use deserialized engine model: /home/shamsee/DeepStream-Yolo/model_b1_gpu0_fp32.engine 0:00:00.412604525 6238 0xaaaad1ab46c0 INFO nvinfer gstnvinfer_impl.cpp:343:notifyLoadModelStatus:<primary_gie> [UID 1]: Load new model:/home/shamsee/DeepStream-Yolo/config_infer_primary_yolo26_1usbcam_flir.txt sucessfully 0:00:00.414435142 6238 0xaaaad1cb8920 WARN basesrc gstbasesrc.c:3127:gst_base_src_loop:<src_elem> error: Internal data stream error. 0:00:00.414475340 6238 0xaaaad1cb8920 WARN basesrc gstbasesrc.c:3127:gst_base_src_loop:<src_elem> error: streaming stopped, reason not-negotiated (-4) 0:00:00.414521843 6238 0xaaaad1ab46c0 WARN v4l2src gstv4l2src.c:794:gst_v4l2src_query:<src_elem> Can't give latency since framerate isn't fixated ! nvstreammux: Successfully handled EOS for source_id=0
Runtime commands: h: Print this help q: Quit
p: Pause r: ResumeNOTE: To expand a source in the 2D tiled display and view object details, left-click on the source. To go back to the tiled display, right-click anywhere on the window.
** INFO: <bus_callback:291>: Pipeline ready
ERROR from src_elem: Internal data stream error. Debug info: ../libs/gst/base/gstbasesrc.c(3127): gst_base_src_loop (): /GstPipeline:pipeline/GstBin:multi_src_bin/GstBin:src_sub_bin0/GstV4l2Src:src_elem: streaming stopped, reason not-negotiated (-4) ** INFO: <bus_callback:256>: incorrect camera parameters provided, please provide supported resolution and frame rate
** INFO: <bus_callback:277>: Pipeline running
** INFO: <bus_callback:334>: Received EOS. Exiting ...
Quitting 0:00:01.483704521 6238 0xaaaad1ab46c0 ERROR GST_BUS gstbus.c:1075:gst_bus_remove_watch: no bus watch was present App run failed
Glenn JocherUltralyticsThat
not-negotiated (-4)is coming fromv4l2srcnot being able to match the caps DeepStream requested (your snippet also has a typo: it must be[source0], and you needtype,camera-width, andcamera-heightset explicitly); please try a minimal V4L2 source config like this (and set[streammux] live-source=1for a USB cam), per the DeepStreamdeepstream-appsource-group keys in the NVIDIA reference app docs and our YOLO26 DeepStream Jetson guide:[source0] enable=1 type=1 camera-v4l2-dev-node=0 camera-width=80 camera-height=60 camera-fps-n=9 camera-fps-d=1 video-format=NV12If it still fails, first confirm GStreamer can negotiate your camera at
80x60@9inUYVY(then we can mirror that into DeepStream), and paste your full[source0]+[streammux]blocks here in this discussion along with the complete text log (not a screenshot).Saarulog
Thanks Glenn. Its working Sir!!!!
Saarulog(已编辑)
Hello I finally got everything else working but stuck in terms of wanting to extract deepstream metadata(inference results-frame, trackerID, detection class, bounding box and etc) and to allow my crewai agents to access for further processing/instructions. I Don't know where to start or what to create(config files or edit files), properties needed or mapping it within the deepstream_app_config.txt/config_infer_primary_yolo26 besides the custom crewai tool(python) to retrieve the needed metadata. I don't want to use cloud storage but locally to jetson orin nano. Any suggestion, direction PLEASE would be GREATLY appreciated. Thanks again for your help and time. SA
Glenn JocherUltralyticsDeepStream already attaches detections/tracks as
NvDsBatchMeta(frame +NvDsObjectMetawithclass_id, bbox, andobject_idoncenvtrackeris enabled), butdeepstream-appisn’t really meant to be “extended” from Python—if you want to feed results into CrewAI locally on the Orin, the usual path is to build a small custom DeepStream app using the NVIDIA DeepStream Python sample apps and add a pad-probe on thenvinfer/nvtrackersrc pad to serialize metadata (e.g., JSON) and hand it directly to your agents.# inside a pad-probe callback import pyds batch_meta = pyds.gst_buffer_get_nvds_batch_meta(hash(gst_buffer)) l_frame = batch_meta.frame_meta_list while l_frame: frame_meta = pyds.NvDsFrameMeta.cast(l_frame.data) l_obj = frame_meta.obj_meta_list while l_obj: obj = pyds.NvDsObjectMeta.cast(l_obj.data) tid = int(obj.object_id) # tracker ID (requires nvtracker) cid = int(obj.class_id) # class r = obj.rect_params # bbox (left, top, width, height) conf = float(obj.confidence) # -> push to your local queue / file / socket for CrewAI l_obj = l_obj.next l_frame = l_frame.nextIf you reply with your JetPack/DeepStream versions and whether you’re currently running the stock
deepstream-appconfig or a custom GStreamer pipeline, I can point you to the closest matching sample to start from.Saarulog(已编辑)
Thanks Glen. I am using Jetpack 6.2.1 and DeepStream 7.1. Deepstream-Yolo(deepstream_app_config.txt). How would the "deepstream_app_config.txt/config_infer_primary_yolo26" be mapped or linked to the deepstream python example you provided above?
Glenn JocherUltralyticsdeepstream_app_config.txtis parsed only bydeepstream-app; the Python samples don’t “map” it automatically, so in Python you recreate the pipeline in code and simply reuse the samenvinferconfig by pointingnvinferat your existingconfig_infer_primary_yolo26.txt(which can keep all your DeepStream-Yolo settings like custom parser lib paths) as shown in the DeepStream on Jetson guide and the upstream DeepStream-Yolo project.# after you create the elements in your Python pipeline pgie = Gst.ElementFactory.make("nvinfer", "primary-gie") pgie.set_property("config-file-path", "config_infer_primary_yolo26.txt") tracker = Gst.ElementFactory.make("nvtracker", "tracker") # set tracker properties from your tracker config the same way (or load them via configparser)
AfaqKarim(已编辑)
Title: Guidance Needed: Where to Modify Code to Add Custom Violation Logic and Modbus Output in DeepStream-YOLO Pipeline
Hello DeepStream-YOLO Team,
First of all, thank you for the excellent DeepStream-YOLO integration repository. I was able to successfully deploy a YOLO26 model using DeepStream on a Jetson Orin Nano (JetPack 6.1), and inference is working correctly using
deepstream-appwithconfig_infer_primary_yolo26.txt.
My Use Case
I am building an industrial safety system. After YOLO detection, if there is violation i have to send 0 or 1 through modbus to the plc: .
So I need to integrate custom logic after detection results.
My Question
In this DeepStream-YOLO setup, in which file or location should I add my custom logic?
Specifically, I want to know:
-
Should I modify:
nvdsparsebbox_Yolo.cppinsidenvdsinfer_custom_impl_Yolo?- Or write a custom DeepStream app (like
deepstream-test1/deepstream-test3)? - Or add a probe function after
nvinferin the pipeline?
Because I understand that:
- Parsing
.sois only for decoding bounding boxes, - But my requirement is to process detections and send a Modbus signal continuously.
So I want guidance on the recommended DeepStream architecture for this.
My Current Setup
- Jetson Orin Nano – JetPack 6.1
- DeepStream-YOLO repo working
- YOLO26 model inference OK
- Using
deepstream-app -c deepstream_app_config.txt
Goal
After detection → evaluate violation → send Modbus 0/1 signal in real-time.
I would really appreciate guidance on:
- Best file/place to implement this logic,
- Any example for accessing detection metadata in DeepStream-YOLO.
Thank you again for your great work 🙏
Glenn JocherUltralyticsYou generally shouldn’t put violation/Modbus logic inside
nvdsparsebbox_Yolo.cpp(keep the custom parser.sofocused on decoding tensors → boxes); the recommended DeepStream architecture is to write a small custom DeepStream app (C/C++ or Python) and attach a pad-probe afternvinfer(or afternvtrackerif you want stable IDs) to readNvDsObjectMetaand then publish your Modbus 0/1 in a non-blocking way (e.g., queue + worker thread), while still following the config/export flow in our Jetson DeepStream + TensorRT guide and the marcoslucianops/DeepStream-Yolo repository.// Probe callback (attach to a pad after nvinfer/nvtracker, e.g. OSD sink pad) static GstPadProbeReturn probe_cb(GstPad* pad, GstPadProbeInfo* info, gpointer user_data) { GstBuffer* buf = (GstBuffer*)info->data; NvDsBatchMeta* batch = gst_buffer_get_nvds_batch_meta(buf); for (NvDsMetaList* lf = batch->frame_meta_list; lf; lf = lf->next) { NvDsFrameMeta* f = (NvDsFrameMeta*)lf->data; for (NvDsMetaList* lo = f->obj_meta_list; lo; lo = lo->next) { NvDsObjectMeta* o = (NvDsObjectMeta*)lo->data; // o->class_id, o->confidence, o->rect_params; compute “violation” then enqueue Modbus 0/1 } } return GST_PAD_PROBE_OK; }If you share whether you need “violation” per-frame or latched with hysteresis (and whether you’re using
nvtracker), reply here and we can suggest the cleanest probe point and metadata fields to use.AfaqKarim
Hi @glenn-jocher — thanks, that makes sense.
My requirements:
Input: RTSP (H.264) on Jetson Orin Nano (JP 6.1 / DS 7.1)
Throughput: ~25 FPS
Tracking: Not using nvtracker (I only need a global safety signal, not per-person IDs)
Decision: Per-frame violation, but I’m open to latching / hysteresis if that’s the recommended way to avoid flicker.
Output: Modbus TCP to PLC; I need to continuously send 0/1 (currently planning to write a coil bit), e.g. modbus_write_bit(coil_addr, value) at ~25Hz (or “send on change” if better).
Violation rule (current):
I define a polygon zone (danger/no-go area).
Violation = 1 if a detected person bbox is fully inside the polygon (all 4 corners inside).
Violation = 0 otherwise.
Questions to confirm best architecture:
Probe location: since I’m not using tracker, is the best place a pad-probe on nvinfer src pad (or OSD sink pad)?
Metadata fields: I’m reading from NvDsObjectMeta → class_id, confidence, rect_params (left/top/width/height). Is that the correct/standard way?
Best practice for Modbus: should I publish every frame (25Hz), or only on state change with a small debounce/hysteresis window (e.g., require N consecutive frames to assert violation)?
If you can point to the cleanest DeepStream sample to start from (e.g., deepstream-test3 or deepstream-app custom plugin approach) and the recommended probe pad in that pipeline, I’ll follow that pattern.
Thanks again 🙏
-
Saarulog
Hello, Deepstream Clarity question please.
I notice the Nvidia Deepstream_python_app sample application use python file and one or more configuration text files. While Ultralytics Deepstream_Yolo uses two configuration text files.
-
Could you please explain why Deepstream_Yolo is different?
-
This command for Yolo26 "python3 export_yolo26.py -w yolo26s.pt" does not seem to generate a working .onnx model file but works with previous Yolo models(Ex. Yolo11). What are my options to get this to work?
Thank you for your help and time.
Glenn JocherUltralytics(已编辑)DeepStream-Yolo isn’t fundamentally “different” from
deepstream_python_apps—it just splits configuration the waydeepstream-appexpects:deepstream_app_config.txtdescribes the full pipeline/sources/sinks, whileconfig_infer_primary_yolo26.txtis thenvinfer(GIE) config that points to the ONNX/engine and the custom YOLO parser library; the Python sample builds more of the pipeline in code so you’ll see fewer/different config files even though the same DeepStream components are in play (see the DeepStream on Jetson + Ultralytics YOLO26 guide).For the YOLO26 ONNX export, it almost always comes down to a JetPack/DeepStream compatibility mismatch with
--dynamic/--opset(DeepStream 6.0.1 and earlier needopset=12, while newer DeepStream uses the exports shown in the guide), so please paste the full terminal output from the export command (text, not a screenshot) and youryolo checksoutput so we can see your exact JetPack/DeepStream/Python/Ultralytics versions (also double-check you followed the Jetson setup in the Jetson quickstart).Saarulog
Hello. Thank you for the quick response. below is the terminal output.
Starting: yolo26s.pt Opening YOLO26 model YOLO26s summary (fused): 122 layers, 9,496,140 parameters, 0 gradients, 20.7 GFLOPs Creating labels.txt file Exporting the model to ONNX W0205 20:50:46.390000 684277 torch/onnx/_internal/exporter/_compat.py:125] Setting ONNX exporter to use operator set version 18 because the requested opset_version 17 is a lower version than we have implementations for. Automatic version conversion will be performed, which may not be successful at converting to the requested version. If version conversion is unsuccessful, the opset version of the exported model will be kept at 18. Please consider setting opset_version >=18 to leverage latest ONNX features W0205 20:50:48.526000 684277 torch/onnx/_internal/exporter/_schemas.py:455] Missing annotation for parameter 'input' from (input, boxes, output_size: 'Sequence[int]', spatial_scale: 'float' = 1.0, sampling_ratio: 'int' = -1, aligned: 'bool' = False). Treating as an Input. W0205 20:50:48.527000 684277 torch/onnx/_internal/exporter/_schemas.py:455] Missing annotation for parameter 'boxes' from (input, boxes, output_size: 'Sequence[int]', spatial_scale: 'float' = 1.0, sampling_ratio: 'int' = -1, aligned: 'bool' = False). Treating as an Input. W0205 20:50:48.528000 684277 torch/onnx/_internal/exporter/_schemas.py:455] Missing annotation for parameter 'input' from (input, boxes, output_size: 'Sequence[int]', spatial_scale: 'float' = 1.0). Treating as an Input. W0205 20:50:48.528000 684277 torch/onnx/_internal/exporter/_schemas.py:455] Missing annotation for parameter 'boxes' from (input, boxes, output_size: 'Sequence[int]', spatial_scale: 'float' = 1.0). Treating as an Input. The model version conversion is not supported by the onnxscript version converter and fallback is enabled. The model will be converted using the onnx C API (target version: 17). Failed to convert the model to the target version 17 using the ONNX C API. The model was not modified Traceback (most recent call last): File "/home/shamsee/Shamsee/lib/python3.10/site-packages/onnxscript/version_converter/init.py", line 127, in call converted_proto = _c_api_utils.call_onnx_api( File "/home/shamsee/Shamsee/lib/python3.10/site-packages/onnxscript/version_converter/_c_api_utils.py", line 65, in call_onnx_api result = func(proto) File "/home/shamsee/Shamsee/lib/python3.10/site-packages/onnxscript/version_converter/init.py", line 122, in _partial_convert_version return onnx.version_converter.convert_version( File "/home/shamsee/Shamsee/lib/python3.10/site-packages/onnx/version_converter.py", line 39, in convert_version converted_model_str = C.convert_version(model_str, target_version) RuntimeError: /github/workspace/onnx/version_converter/BaseConverter.h:65: adapter_lookup: Assertion
falsefailed: No Adapter To Version $17 for Resize Done: yolo26s.onnx
Glenn JocherUltralytics(已编辑)That log shows PyTorch is actually exporting at opset 18 and then failing to down-convert to opset 17 (
No Adapter To Version 17 for Resize), so the fix is to export with an opset that matches what the exporter is using (>=18) instead of forcing 17:python3 export_yolo26.py -w yolo26s.pt --opset 18If you’re on an older DeepStream/TensorRT stack that truly requires lower opset (e.g. DeepStream 6.0.1), you’ll need the DeepStream 6.0.1 and earlier setting from the guide (
opset=12) and typically an older PyTorch ONNX exporter that can produce opset 12 cleanly; please paste youryolo checksoutput and the exact DeepStreamnvinferONNX parsing error so we can align this to your JetPack/DeepStream version (see the DeepStream on Jetson + Ultralytics YOLO26 guide and the Jetson quickstart).Saarulog(已编辑)
smh, User Error. :-) I forgot about the steps at https://docs.ultralytics.com/guides/nvidia-jetson/ . It works after following steps!!!
One last issue for Segmentation ONNX conversion. Hopefully this is another user error. python3 export_yolo26.py -w yolo26s-seg.pt
I got these error:
Starting: yolo26s-seg.pt Opening YOLO26 model YOLO26s-seg summary (fused): 139 layers, 10,396,300 parameters, 0 gradients, 34.2 GFLOPs Creating labels.txt file Exporting the model to ONNX Traceback (most recent call last): File "/home/shamsee/ultralytics/export_yolo26.py", line 157, in main(args) File "/home/shamsee/ultralytics/export_yolo26.py", line 116, in main torch.onnx.export( File "/home/shamsee/Sentinel_yolo26/Sentinel_proj/lib/python3.10/site-packages/torch/onnx/utils.py", line 549, in export _export( File "/home/shamsee/Sentinel_yolo26/Sentinel_proj/lib/python3.10/site-packages/torch/onnx/utils.py", line 1641, in _export graph, params_dict, torch_out = _model_to_graph( File "/home/shamsee/Sentinel_yolo26/Sentinel_proj/lib/python3.10/site-packages/torch/onnx/utils.py", line 1184, in _model_to_graph graph, params, torch_out, module = _create_jit_graph(model, args) File "/home/shamsee/Sentinel_yolo26/Sentinel_proj/lib/python3.10/site-packages/torch/onnx/utils.py", line 1060, in _create_jit_graph graph, torch_out = _trace_and_get_graph_from_model(model, args) File "/home/shamsee/Sentinel_yolo26/Sentinel_proj/lib/python3.10/site-packages/torch/onnx/utils.py", line 964, in _trace_and_get_graph_from_model trace_graph, torch_out, inputs_states = torch.jit._get_trace_graph( File "/home/shamsee/Sentinel_yolo26/Sentinel_proj/lib/python3.10/site-packages/torch/jit/_trace.py", line 1502, in _get_trace_graph outs = ONNXTracedModule( File "/home/shamsee/Sentinel_yolo26/Sentinel_proj/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1714, in _wrapped_call_impl return self._call_impl(*args, **kwargs) File "/home/shamsee/Sentinel_yolo26/Sentinel_proj/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1725, in _call_impl return forward_call(*args, **kwargs) File "/home/shamsee/Sentinel_yolo26/Sentinel_proj/lib/python3.10/site-packages/torch/jit/_trace.py", line 141, in forward graph, out = torch._C._create_graph_by_tracing( File "/home/shamsee/Sentinel_yolo26/Sentinel_proj/lib/python3.10/site-packages/torch/jit/_trace.py", line 132, in wrapper outs.append(self.inner(*trace_inputs)) File "/home/shamsee/Sentinel_yolo26/Sentinel_proj/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1714, in _wrapped_call_impl return self._call_impl(*args, **kwargs) File "/home/shamsee/Sentinel_yolo26/Sentinel_proj/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1725, in _call_impl return forward_call(*args, **kwargs) File "/home/shamsee/Sentinel_yolo26/Sentinel_proj/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1704, in _slow_forward result = self.forward(*input, **kwargs) File "/home/shamsee/Sentinel_yolo26/Sentinel_proj/lib/python3.10/site-packages/torch/nn/modules/container.py", line 249, in forward input = module(input) File "/home/shamsee/Sentinel_yolo26/Sentinel_proj/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1714, in _wrapped_call_impl return self._call_impl(*args, **kwargs) File "/home/shamsee/Sentinel_yolo26/Sentinel_proj/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1725, in _call_impl return forward_call(*args, **kwargs) File "/home/shamsee/Sentinel_yolo26/Sentinel_proj/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1704, in _slow_forward result = self.forward(*input, **kwargs) File "/home/shamsee/ultralytics/export_yolo26.py", line 34, in forward x = x.transpose(1, 2) AttributeError: 'tuple' object has no attribute 'transpose'
Thanks for your help and time.
Glenn JocherUltralyticsThat traceback is expected with
yolo26*-seg: theexport_yolo26.pyyou’re using (from DeepStream-Yolo) is built around detection-only outputs, while segmentation models return multiple outputs (so you end up with a tuple and hit thetransposefailure); for DeepStream segmentation please use the dedicated DeepStream-Yolo-Seg project (its supported-model list currently covers YOLO11/YOLOv8 seg), and if you specifically need YOLO26-Seg in DeepStream please open a report in DeepStream-Yolo-Seg Issues with your JetPack/DeepStream versions plusyolo checksso it’s trackable.
-
lovechang1986
I'm working on integrating SAHI with YOLOv11 and deploying it on a Jetson device using DeepStream. The original image size is 14401080, and I plan to split it into 640640 sub-images with overlaps between different sub-images. Are there any good methods for this?
Additionally, I still hope to use DeepSort for inference and tracking after obtaining the final image.
Glenn JocherUltralyticsSAHI-style slicing isn’t something DeepStream handles automatically, so you’ll either need to slice frames before they enter the DeepStream pipeline (then offset + merge all tile detections back into full-frame coordinates before tracking), or implement ROI/tiling as a DeepStream preprocess step (e.g., via
nvdspreprocessor a small custom plugin) so the primary GIE sees a batched set of crops; on Jetson it’s often simpler/faster to avoid slicing and instead run a single-pass TensorRT engine at a largerimgszif latency allows (and YOLO26 is the recommended starting point for new deployments). If you plan to track inside DeepStream, put the tracker after you’ve merged detections to full-frame coordinates; DeepStream’s built-innvtrackerworks out of the box, while DeepSORT specifically typically requires custom integration rather than a drop-in switch. If you paste your JetPack/DeepStream versions and your currentdeepstream_app_config.txtinto this discussion, we can suggest the cleanest path based on the exact pipeline; the baseline export + DeepStream wiring is in the Jetson DeepStream guide and Jetson quickstart.
metalpole-coder
Do you know how to get the detections to work within a ROI only in deepstream?