Feature Selection for Vision based Grasping

Research Project Goal

The idea is to apply feature selection techniques to improve performance of the grasping system. (One that labels each point in the image as a good grasp or not.) One may use 3-d data as well.

Research Scope

Use machine learning techniques for determine good features for grasping.

Sample Data

• Data will be made available by the mentor.

Tasks

Challenges

Research Project Status

student names here

Point of Contact

Ashutosh Saxena (asaxena@cs.stanford.edu)

Midterm Report

not yet submitted

Final Report

not yet submitted