SLAM結果評測工具——TUM Banchmark & EVO

1.TUM評估工具：

absolute trajectory error (ATE) and the relative pose error (RPE)

python associate.py rgb.txt depth.txt > associate.txt

python generate_pointcloud.py rgb.png depth.png  res.ply

ATE：

python evaluate_ate.py groundtruth.txt estimated.txt --plot PLOT

python evaluate_ate.py groundtruth.txt estimated.txt --max_pairs 10000 --plot figure.png --offset 0 --scale 1 --verbose

RPE：

python evaluate_rpe.py groundtruth.txt estimated.txt --fixed_delta --plot figure.png --offset 0 --scale 1 --verbose

參考：
https://vision.in.tum.de/data/datasets/rgbd-dataset/tools#evaluation
https://www.cnblogs.com/ilym/p/9525484.html
https://blog.csdn.net/bingoplus/article/details/73521955

2.EVO評估工具：

evo_ape - absolute pose error (APE) and evo_rpe - relative pose error (RPE)
安裝：
pip install evo --upgrade --no-binary evo

pip install evo --upgrade --no-binary evo --user

使用：

cd /home/hong/Desktop/drawtrajectory/EVO

評測軌跡：

evo_traj tum groundtruth.txt estimated.txt --ref groundtruth.txt -va -p 
--plot_mode=xz

evo_ape - absolute pose error：

evo_ape tum groundtruth.txt estimated.txt  -va -p 

evo_rpe - relative pose error：

(保存結果爲ZIP文件)

mkdir results 

evo_rpe tum groundtruth.txt estimated.txt  -va -p --save_results results/result1.ZIP

將ZIP結果存爲table.txt

evo_res result/result1.ZIP -p --save_table results/table.txt

evo_traj kitti KITTI_00_ORB.txt KITTI_00_SPTAM.txt --ref=KITTI_00_gt.txt -p --plot_mode=xz

不同格式的軌跡文件及相互轉換：
Examples：

evo_traj euroc data.csv --save_as_tum

https://github.com/MichaelGrupp/evo/wiki/Formats

注：
生成的文件中偶爾存在空格等不是完全規範的TUM結果，執行下列命令進行清除多餘的空格：

cat results.txt | tr -s [:space:] > results_new.txt

參考：
https://github.com/MichaelGrupp/evo
https://blog.csdn.net/weixin_44386661/article/details/103080551
https://blog.csdn.net/A_L_A_N/article/details/88708979
https://www.e-learn.cn/content/qita/2521778
http://www.liuxiao.org/2017/11/%E4%BD%BF%E7%94%A8-evo-%E5%B7%A5%E5%85%B7%E8%AF%84%E6%B5%8B-vi-orb-slam2-%E5%9C%A8-euroc-%E4%B8%8A%E7%9A%84%E7%BB%93%E6%9E%9C/
https://zhuanlan.zhihu.com/p/48231990
https://zhuanlan.zhihu.com/p/88223106
https://blog.csdn.net/CSDNhuaong/article/details/101909888

3.Linux安裝MeshLab

PPA 安裝，打開終端，輸入以下命令：
sudo add-apt-repository ppa:zarquon42/meshlab
sudo apt-get update
sudo apt-get install meshlab

卸載命令：
sudo apt-get remove meshlab

移除PPA，輸入以下命令：
sudo apt-get install ppa-purge
sudu ppa-purge ppa:zarquon42/meshlab

參考： https://www.wandouip.com/t5i213024/