Copyright (C) 2026 Computer Vision Lab, Electrical Engineering, 
Indian Institute of Science, Bengaluru, India.
All rights reserved.

This file is a part of the implementation for the paper:
Lalit Manam and Venu Madhav Govindu, Sensitivity Analysis in Translation Averaging, IEEE Transactions on Pattern Analysis and Machine Intelligence, 2026.

This folder contains demo code and data for removing ill-conditioned nodes from a bearing-based network. 
The code is tested on Matlab 2023a.

Main file to run the code: Demo_Filter_Nodes.m.
Demo Data: demo_data.mat.
It contains the edges list and relative directions all placed in a common reference frame. Refer to Lines 36 to 40 in Demo_Filter_Nodes.m for the structure of the variables.

parfor is used for fast computation of conditioning of the graph and parallelization of the algorithm (dependency: Parallel Computing Toolbox). To avoid using parfor, set parallelVer to False in line 48 of Demo_Filter_Nodes.m.
