Detecting Expertise in Gaze Data Using Hidden-Markov-Models

{"The":[0],"circular":[1],"factory":[2],"concept":[3],"offers":[4],"a":[5,26,115],"promising":[6],"solution":[7],"to":[8,70,80,90,125,138,149,152,187],"rising":[9],"global":[10],"resource":[11],"demand.":[12],"To":[13,64],"realize":[14],"this,":[15],"the":[16,30,47,81,126,189],"Collaborative":[17],"Research":[18],"Centre":[19],"(CRC)":[20],"1574":[21],"\\"Circular":[22],"Factory\\"":[23],"is":[24,43],"developing":[25],"lab-scale":[27],"setup":[28],"where":[29],"disassembly":[31,105],"and":[32,56,106,136,160,195],"reassembly":[33,107],"of":[34,46,49,108,181],"angle":[35,109],"grinders":[36],"are":[37,147,163],"central":[38],"tasks.":[39],"Automating":[40],"this":[41,67],"process":[42],"complex,":[44],"because":[45],"nature":[48],"used":[50],"products,":[51],"which":[52],"vary":[53],"in":[54],"generation":[55],"condition.":[57],"Therefore,":[58],"human":[59,92,192],"expertise":[60,93,155,172],"currently":[61],"remains":[62],"essential.":[63],"successfully":[65],"transfer":[66],"intrinsic":[68],"knowledge":[69],"robots":[71],"via":[72,114],"learning-from-demonstration,":[73],"training":[74,162],"data":[75,113,158],"must":[76],"be":[77],"weighted":[78],"according":[79],"demonstrator's":[82],"expertise.":[83],"This":[84,185],"work":[85],"proposes":[86],"an":[87,98],"automated":[88,179],"approach":[89],"evaluate":[91],"using":[94],"eye-tracking.":[95],"We":[96,165],"conducted":[97],"experiment":[99],"(N":[100],"=":[101],"38)":[102],"involving":[103],"manual":[104],"grinders,":[110],"collecting":[111],"gaze":[112,141,176],"Tobii":[116],"Pro":[117],"Glasses":[118],"3":[119],"system.":[120],"Gaze":[121],"points":[122],"were":[123],"mapped":[124],"five":[127],"distinct":[128],"Areas-of-Interest":[129],"(AOIs)":[130],"‘Instructions’,":[131],"‘Bins’,":[132],"‘Tools’,":[133],"‘Angle":[134],"Grinder’,":[135],"‘Other’":[137],"generate":[139],"discrete":[140],"sequences.":[142],"Hidden":[143],"Markov":[144],"Models":[145],"(HMMs)":[146],"applied":[148],"these":[150],"sequences":[151],"classify":[153],"participants'":[154],"levels.":[156],"Currently,":[157],"analysis":[159],"HMM":[161],"underway.":[164],"hypothesize":[166],"that":[167],"HMMs":[168],"can":[169],"effectively":[170],"distinguish":[171],"levels":[173],"based":[174],"on":[175],"patterns,":[177],"enabling":[178],"selection":[180],"high-quality":[182],"demonstration":[183],"data.":[184],"contributes":[186],"bridging":[188],"gap":[190],"between":[191],"cognitive":[193],"skills":[194],"highly":[196],"adaptable":[197],"robotic":[198],"automation.":[199]}

Authors

Publication Details

Journal
KITopen
Published
2026-09-17
DOI
https://doi.org/10.5445/ir/1000197047
Primary Topic
Gaze Tracking and Assistive Technology
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
article

Detecting Expertise in Gaze Data Using Hidden-Markov-Models

Manuel Zaremski, Linus Kunzmann, Barbara Deml
KITopen
Gaze Tracking and Assistive Technology
article

Detecting Expertise in Gaze Data Using Hidden-Markov-Models

Manuel Zaremski, Linus Kunzmann, Barbara Deml
article en

Abstract

The circular factory concept offers a promising solution to rising global resource demand. To realize this, the Collaborative Research Centre (CRC) 1574 "Circular Factory" is developing a lab-scale setup where the disassembly and reassembly of angle grinders are central tasks. Automating this process is complex, because of the nature of used products, which vary in generation and condition. Therefore, human expertise currently remains essential. To successfully transfer this intrinsic knowledge to robots via learning-from-demonstration, training data must be weighted according to the demonstrator's expertise. This work proposes an automated approach to evaluate human expertise using eye-tracking. We conducted an experiment (N = 38) involving manual disassembly and reassembly of angle grinders, collecting gaze data via a Tobii Pro Glasses 3 system. Gaze points were mapped to the five distinct Areas-of-Interest (AOIs) ‘Instructions’, ‘Bins’, ‘Tools’, ‘Angle Grinder’, and ‘Other’ to generate discrete gaze sequences. Hidden Markov Models (HMMs) are applied to these sequences to classify participants' expertise levels. Currently, data analysis and HMM training are underway. We hypothesize that HMMs can effectively distinguish expertise levels based on gaze patterns, enabling automated selection of high-quality demonstration data. This contributes to bridging the gap between human cognitive skills and highly adaptable robotic automation.

KITopen
Openalex Percentile: Top 8%
Gaze Tracking and Assistive Technology
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.