Writeup Template: You can use this file as a template for your writeup if you want to submit it as a markdown file, but feel free to use some other method and submit a pdf if you prefer.
First of all, I've used the previous Exercise which is 'sensor_stick' Exercise.
- filtering RANSAC Exercise : https://www.youtube.com/watch?v=vt0QpQmOjto - In this project, passthrough filter should be two. I've realized there are two boxes one the left and right hand side. you should make a filter with y-axis
- Cluster Visulization : https://www.youtube.com/watch?v=xcQ5ySgGAMM
- Object Recognition : https://www.youtube.com/watch?v=bwYcwcsnXbU
- Train SVM : https://www.youtube.com/watch?v=GXwLPqTjkZ8 also contained in this repository
To Train SVM in project. I've used the 'sensor_stick' project. I've changed models and capture features. moving models folder from Project to sensor_stick (models_backup is for sensor_stick) before capturing features, should change models list
Train Project SVM : https://www.youtube.com/watch?v=dOeFrYC3-7Y the accuracy was not that great. perhaps I should've looped more.
copied from model.sav which I'd made before in sensor_stick to project script folder. and recognize.
result outputs is in yaml folder and images are below. yaml_01 yaml_02 yaml_03
changed several variables and I have more accurate model.sav second training.sav second model.sav
REVIEW:
- To finish this project. what I could do was just reading. I've never learned like this. so, I sometimes used LIVE asking and just read lectures over and over.
- To get better result. I could spend more time to capture features( more loop). that will make better model.sav and better result.
============================================================================