Project Goal

The goal is to develop an SFM algorithm using video acquired by an autonomous helicopter. The end goal of this project are (1) to fly helicopters using on-board cameras for position estimation; (2) to reconstruct the ground structure, and (3) to find moving objects on the ground, all open research topics.

---

Figure 1: The Stanford Autonomous Helicopter

---

Project Scope

The project involves the implementation of techniques to detect invariant features known as SIFT features. The SIFT features are to be tracked across multiple images, to determine the structure (model) of the ground as well as the location of the vehicle.

Tasks

The project will be accomplished through the following tasks.

• Task 1: Implement SIFT feature detectors

• Task 2: Track features over multiple camera frames

• Task 3: Perform SFM.

You will be working closely with the Stanford Autonomous Helicopter Lab, which will provide the necessary infrastructure to collect the data and evaluate your results. The Lab will provide model helicopters flown by one of our experienced pilots.

Project Status

Varun Ganapathi, Timothy Lee, Eric Berger

Point of Contact

Sebastian Thrun

Midterm Report

submitted

Final Report

submitted