Exploring the Role of Visual Crowding in Masking Mesh Decimation Artifacts in Virtual Reality.

Immersive head-mounted displays (HMDs) engage a large portion of peripheral vision, where visual crowding imposes a fundamental limitation on object recognition. While crowding is well documented with 2D laboratory stimuli, its consequences for 3D content optimization in virtual reality (VR) remain insufficiently understood. We investigate, in a controlled VR setting, whether visual crowding can act as a masking mechanism for decimation-induced geometric artifacts on convex low-polygon mesh objects. Here, such artifacts refer to visible surface-detail loss, local geometric changes that introduce salient shape irregularities, incoherent contours, and silhouette deviations resulting from edge-collapse. However, the aim was not to explore each of them independently. In a controlled VR psychophysics experiment with eye-tracking enforcement of fixation, observers reported whether they could perceive any visible geometric distortion introduced by decimation on a peripheral low-polygon icosphere target. We systematically manipulated lateral target-flanker spacing (Bouma factor), axial separation through binocular disparity, and the depth offset between the fixation plane and the stimuli. Distortion detection thresholds show robust threshold elevation under crowding: stronger decimation is tolerated when flankers are nearby, and sensitivity progressively recovers with increasing lateral and axial separation, yet crowding remains significant even for large depth separations (beyond \(\sim\) 200 arcsec). Beyond visibility thresholds, crowding also increases decision time: reaction times are systematically longer in crowded configurations than in the uncrowded baseline. Finally, we observe a marked directional anisotropy: threshold elevation is strongest when the fixation point lies below the stimuli (Down fixation), suggesting that measured crowding effects may vary with fixation direction which is consistent with prior findings in vision sciences. Together, these results connect vision-science findings on crowding with the analysis of mesh artifacts in VR. They suggest that local inter-object spacing could help guide future LoD (Level of Detail) selection strategies.

Authors

Institutions

Publication Details

Journal
ACM Transactions on Applied Perception
Published
2026-11-19
DOI
https://doi.org/10.1145/3862767
Primary Topic
Visual perception and processing mechanisms
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Exploring the Role of Visual Crowding in Masking Mesh Decimation Artifacts in Virtual Reality.

Mohamed–Chaker Larabi, Steven Le Moan, Rivo Andriamanalina
ACM Transactions on Applied Perception
Visual perception and processing mechanisms
article

Exploring the Role of Visual Crowding in Masking Mesh Decimation Artifacts in Virtual Reality.

Mohamed–Chaker Larabi, Steven Le Moan, Rivo Andriamanalina
article en

Abstract

Immersive head-mounted displays (HMDs) engage a large portion of peripheral vision, where visual crowding imposes a fundamental limitation on object recognition. While crowding is well documented with 2D laboratory stimuli, its consequences for 3D content optimization in virtual reality (VR) remain insufficiently understood. We investigate, in a controlled VR setting, whether visual crowding can act as a masking mechanism for decimation-induced geometric artifacts on convex low-polygon mesh objects. Here, such artifacts refer to visible surface-detail loss, local geometric changes that introduce salient shape irregularities, incoherent contours, and silhouette deviations resulting from edge-collapse. However, the aim was not to explore each of them independently. In a controlled VR psychophysics experiment with eye-tracking enforcement of fixation, observers reported whether they could perceive any visible geometric distortion introduced by decimation on a peripheral low-polygon icosphere target. We systematically manipulated lateral target-flanker spacing (Bouma factor), axial separation through binocular disparity, and the depth offset between the fixation plane and the stimuli. Distortion detection thresholds show robust threshold elevation under crowding: stronger decimation is tolerated when flankers are nearby, and sensitivity progressively recovers with increasing lateral and axial separation, yet crowding remains significant even for large depth separations (beyond \(\sim\) 200 arcsec). Beyond visibility thresholds, crowding also increases decision time: reaction times are systematically longer in crowded configurations than in the uncrowded baseline. Finally, we observe a marked directional anisotropy: threshold elevation is strongest when the fixation point lies below the stimuli (Down fixation), suggesting that measured crowding effects may vary with fixation direction which is consistent with prior findings in vision sciences. Together, these results connect vision-science findings on crowding with the analysis of mesh artifacts in VR. They suggest that local inter-object spacing could help guide future LoD (Level of Detail) selection strategies.

ACM Transactions on Applied Perception
Centre National de la Recherche Scientifique (FR), Norwegian University of Science and Technology (NO), Université de Poitiers (FR), XLIM (FR), Université de Limoges (FR)
Visual perception and processing mechanisms
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.

Laika access coming soon