How to using jetson inference to training my hand datasets?

lyysupertom · October 30, 2024, 7:53am

I have two problem of this document,
[[jetson-inference/docs/posenet.md at master · dusty-nv/jetson-inference · GitHub](https://github.com/dusty-nv/jetson-inference/blob/master/docs/posenet.md)](https://posenet)
i see the video of hand pose detection that is so unstable and easy to get the keypoint to other place.
i don’t know what reason cause this problem.
is it the problem of dataset or other reason? and i do not see any doc to teach me how to train my hand pose dataset using resNet-18.
i try to follow this projet and get the model to test ,
https://github.com/NVIDIA-AI-IOT/trt_pose_hand
but my realsense camera seem no support to using Jetcam libary.
how to translate the .pth model to onnx to use in Jetson_inference lib. i try to transfer the .pth but faild

import torch
import onnx
import torchvision
# Load  PyTorch model
model = torchvision.models.resnet18(pretrained=True)

# Set  input shape of the model
input_shape = (1, 3, 256, 256)

# Export  PyTorch model to ONNX format
torch.onnx.export(model, torch.randn(input_shape), 'resnet18.onnx', opset_version=11)

could you help me how to training or how to transfer .pth to onnx. i am a newbie. thank you!

AastaLLL · October 30, 2024, 8:31am

Hi,

What kind of error do you meet when converting the model into ONNX?
PyTorch library should support the ONNX export already.

Usually, you will need a desktop GPU to train a model.
Or you can find the below tutorial for the detector:

github.com

dusty-nv/jetson-inference/blob/master/docs/pytorch-transfer-learning.md

<img src="https://github.com/dusty-nv/jetson-inference/raw/master/docs/images/deep-vision-header.jpg" width="100%">
<p align="right"><sup><a href="depthnet.md">Back</a> | <a href="pytorch-cat-dog.md">Next</a> | </sup><a href="../README.md#hello-ai-world"><sup>Contents</sup></a>
<br/>
<sup>Transfer Learning</sup></s></p>

# Transfer Learning with PyTorch

Transfer learning is a technique for re-training a DNN model on a new dataset, which takes less time than training a network from scratch.  With transfer learning, the weights of a pre-trained model are fine-tuned to classify a customized dataset.  In these examples, we'll be using the <a href="https://arxiv.org/abs/1512.03385">ResNet-18</a> and [SSD-Mobilenet](pytorch-ssd.md) networks, although you can experiment with other networks too.

<p align="center"><a href="https://arxiv.org/abs/1512.03385"><img src="https://github.com/dusty-nv/jetson-inference/raw/master/docs/images/pytorch-resnet-18.png" width="600"></a></p>

Although training is typically performed on a PC, server, or cloud instance with discrete GPU(s) due to the often large datasets used and the associated computational demands, by using transfer learning we're able to re-train various networks onboard Jetson to get started with training and deploying our own DNN models.  

<a href=https://pytorch.org/>PyTorch</a> is the machine learning framework that we'll be using, and example datasets along with training scripts are provided to use below, in addition to a camera-based tool for collecting and labeling your own training datasets.  

## Installing PyTorch

If you are [Running the Docker Container](aux-docker.md) or optionally chose to install PyTorch back when you [Built the Project](building-repo-2.md#installing-pytorch), it should already be installed on your Jetson to use.  Otherwise, if you aren't using the container and want to proceed with transfer learning, you can install it now:

``` bash

This file has been truncated. show original

Thanks.

lyysupertom · October 30, 2024, 8:45am

I already read this page and i just curious is there are any dataset for Hand detection i didn’t see. if i want to make dataset for hand keypoint how to make the folder and label through jetson nano and it is the same as sample resnet—18 model ? i didnt see any document to show this kind of dataset. so i dont know how to do this.i only see coco dataset for downloading from this page.by the way may I use jetson nano GPU to train resnet model? thank you

AastaLLL · November 4, 2024, 6:46am

Hi

Another sample will be trt_pose_hand.
You can find some related info in data_collection and gesture_training folder:

Thanks.

Topic		Replies	Views
Jetson Nano - Transfer Learning on Object Detection Model Jetson Nano	13	3178	October 14, 2021
Jetsonnano 2gb custom model live inference? Jetson Nano jetson-inference	8	614	May 6, 2022
Train custom object detectio model Jetson Nano ai-training	12	3005	October 18, 2021
Retraining data for Object Detection Jetson Nano jetson-inference	2	326	October 15, 2021
Question about using jetson-inference in production Jetson TX2 jetson-inference	7	1334	October 18, 2021
Re-trained Pytorch Mask-RCNN inferencing in Jetson Nano Jetson Nano pytorch	2	1552	October 18, 2021
Nano B01 crashes while installing PyTorch Frameworks pytorch	1	1106	June 21, 2020
How to run yolox on Jetson Nano Jetson Nano jetson-inference	6	1125	December 6, 2023
Re-training inception v4 Jetson Nano jetson-inference	4	1038	December 1, 2021
Loading custom Object Detection Model to Jetson Dev for real-time use Jetson Nano jetson-inference	4	577	October 15, 2021

How to using jetson inference to training my hand datasets?

Related topics