Table 1 Failure is a word unknown to me.

From: Large scale analysis of dataset and simulation biases in SLAM research

SLAM

Failure cases

Failure instances

RatSLAM32, 2004

  

MonoSLAM33, 2007

  

PTAM34, 2009

\(\checkmark\) \(\checkmark\) \(\checkmark\)

 

1 pt RANSAC VO35, 2010

  

DTAM36, 2011

\(\checkmark\) \(\checkmark\)

 

LSDSLAM37, 2014

  

SVO38, 2014

  

ORB-SLAM39, 2015

\(\checkmark\)

1

VINS-Mono40, 2018

  

ORB-SLAM341, 2021

\(\checkmark\)

2

MASt3R-SLAM42, 2025

\(\checkmark ^*\)

 
  1. Limited number of failure cases (initialization failure, tracking failure) are reported with SLAM pipelines. Second column indicates the acknowledgment of failure cases. Last column indicates the number of failure instances reported. \(\checkmark ^*\): Recent MASt3R-SLAM does not include any failure cases; they have however discussed limitations of their approach.
Back to article page