/cp-net

object pose estimation using neural network

Primary LanguagePython

cp-net

object pose estimation using neural network

package explanation

cp_net

pose estimation using deep neural network

Data Set

linemod dataset

requirements

Eigen, cython, chainer v3, cuda, scikit-learn, cv2, etc...

https://github.com/oshiroy/pose_estimation_gpu

Sample

pip install -e .
mkdir train_data
cd train_data
## download train_data from (https://drive.google.com/file/d/1A2skh4mYgsoXXKnLOuQ2BVPsnSoe8PZu/view?usp=sharing) in train_data
tar xfvz linemodSIXD2017.tar.gz
cd ..

mkdir trained_data
cd trained_data
## download network model from (https://drive.google.com/file/d/1fiBsCbx0kWK76lhZZ-EGDrRJ35hFBNsu/view?usp=sharing) in trained_data
tar xfvz model_iteration-100000.tar.gz
cd ..

# demo
cd examples/linemodSIXD2017
cd python demo.py -g 0 ../../trained_data/model_iteration-100000