Marcos Luciano

icon indicating a website link https://www.linkedin.com/in/marcoslucianops

icon location Brazil icon location Sr. Computer Vision Engineer | Control and Automation, Mechatronics Engineer

Results 468 comments of Marcos Luciano

Does this code support yolo-pose？

It doesn't support pose yet.

Does this code support yolo-pose？

The pose estimation needs a custom output layer using the TensorMeta. It's more complex.

Does this code support yolo-pose？

Added support for Pose: https://github.com/marcoslucianops/DeepStream-Yolo-Pose

config_tracker_NvDeepSORT.yml dose not work

https://docs.nvidia.com/metropolis/deepstream/dev-guide/text/DS_plugin_gst-nvtracker.html#setup-and-usage-of-re-id-model

Introduction of ONNX, questions Yolo::buildYoloNetwork existence

The `Yolo::buildYoloNetwork` is for Darknet models (YOLOv2, YOLOv3 and YOLOv4).

Deepstream upgrade

it's not possible. You can use the same exported ONNX model to generate the engine file.

Feature Request: Add support for yoloV8 Instance Segmentation models

Yes, soon.

Feature Request: Add support for yoloV8 Instance Segmentation models

https://github.com/marcoslucianops/DeepStream-Yolo-Seg

onnx from T4 to G4

The ONNX model yes, but you need to generate the engine again running the `deepstream-app` command (or running a custom DeepStream code).

Inference video does not have an inference box

Are there any bbox on the detection? Are you testing with the `deepstream-app`?