IndusGrasp

KO | EN

gitlite — search

#llm #golang #go #machine-learning #linux #docker #windows #cli #vue #awesome-list #security #pytorch #openai #open-source

(★ 41)

This is the official implementation of our paper: "A Novel Robotic Grasp Detection Framework Using Low-Cost RGB-D Camera for Industrial Bin Picking."

IndusGrasp 최신버젼 다운로드

최종 버전 다운로드 (.zip)

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# IndusGrasp
This is the official PyTorch Implementation of **IndusGrasp: A Novel Robotic Grasp Detection Method Using Synthetic
Data for Disordered Industrial Scenarios**

## Video
- The video is available in [YouTube](https://youtu.be/lmlCMYdMw5g)

## The pipeline:

The Comparison studies in the real-world environment
<img src="image/compare.png" width="950" height="240"/>

## Requirements: Software
  ```shell
cd bop_toolkit

pip install -r requirements.txt -e .

sudo apt-get install libglfw3-dev libglfw3

sudo apt-get install libassimp-dev

pip install  cython cyglfw3 pyassimp==3.3 imgaug progressbar

pip install  -r requments.txt
  ```

## Generate dataset
- Using the `tools/pybullet_dataset/pybullet_dataset.py` to create render poses with your own 3D mesh(xx.urdf)

- Using the `tools/creat_grasp.py` to create grasp representation with your own 3D mesh(xx.ply)

- Using the `dataset_generate.py` to create your dataset

## Train the IndusGrasp
`python train.py`

## Test the IndusGrasp
`python test_with_detection.py ` use the maskrcnn in torchvision to detect the object firstly.
`python test_without_detection.py ` just detect the grasp pose.

## Comparison Studies
- Our Method  in dense scenarios

- GR-CNN, in sparse scenarios

- GraspNet, which doesn't work for industrial parts

## Generalization Studies
- Our method can generate to unseen objects with different shapes and textures
(Note: the network is only trained by Synthetic Data generated by object 1. )

## Notice
This version use the pyrender (from [bop_toolkit](https://github.com/thodan/bop_toolkit)) to replace the opengl.
There maybe some problems in the maskrcnn train/test with cuda, but this model is 
small, and you can run it with cpu. In addition, this the grasp sampling method of this project is not the best solution.
For better performance, you can use this [easy-dexnet_vis](https://github.com/sunhan1997/easy-dexnet_vis) to sample grasps.

IndusGrasp

IndusGrasp 최신버젼 다운로드

오늘의 추천 프로젝트

Was this content helpful?

Recent Feedback

IndusGrasp

IndusGrasp 최신버젼 다운로드

오늘의 추천 프로젝트

Was this content helpful?

Recent Feedback

Download README