J. D. Hilger

ORCID: 0000-0003-1458-4927
Publications
Citations
Views
---
Saved
---
About
Contact & Profiles
Research Areas
  • Advanced Graph Neural Networks
  • Computer Graphics and Visualization Techniques
  • Surface Roughness and Optical Measurements
  • Visual perception and processing mechanisms
  • Advanced Neuroimaging Techniques and Applications
  • Advanced Image Processing Techniques
  • Caching and Content Delivery
  • Peer-to-Peer Network Technologies
  • Advanced Vision and Imaging
  • Ophthalmology and Visual Impairment Studies
  • Spatial Neglect and Hemispheric Dysfunction
  • Industrial Vision Systems and Defect Detection
  • Mental Health via Writing
  • Advanced Numerical Analysis Techniques
  • Recommender Systems and Techniques

University of Luxembourg
2023-2025

University of California, Los Angeles
2010

UCLA Health
2004

Problem: Contour interpolation has been shown to be scale-invariant in some respects, as the effect of support ratio on contour strength. Kellman & Shipley (1991) hypothesized that mathematical criteria relatability included a small tolerance for misaligned parallel edges. Empirical estimates have suggested collinear edges can up about 15 arc min and still interpolation. We investigated whether retinal metric or notion, such misalignment edge separation, could account data. Methods:...

10.1167/5.8.571 article EN cc-by-nc-nd Journal of Vision 2010-03-17

Graph related tasks, such as graph classification and clustering, have been substantially improved with the advent of neural networks (GNNs). However, existing embedding models focus on homogeneous graphs that ignore heterogeneity graphs. Therefore, using heterogeneous discards rich semantics achieves average performance, especially by utilizing unlabeled information. limited work has done whole a supervised task. In light this, we investigate unsupervised distributed representations...

10.1145/3555776.3577740 article EN Proceedings of the 37th ACM/SIGAPP Symposium on Applied Computing 2023-03-27

Although the role of surface-level processes has been demonstrated, models visual interpolation emphasize contour relationships. We report research on geometric constraints governing 3D between surface patches having no visible contours. In a previous study, we demonstrated that relatability acts as cue for spatial unification in absence explicit edge information. Observers were asked to classify pairs planar — specified by random dot disparities and through circular apertures...

10.1167/6.6.334 article EN cc-by-nc-nd Journal of Vision 2010-03-18

Models of visual interpolation emphasize contour relationships. Although the role surface-level processes has been demonstrated (Yin, Kellman & Shipley, 1998; Fantoni, Bertamini Gerbino, 2004; for a recent review see Kellman, 2003), specific surface properties and geometric constraints that govern are not well understood. In this study we hypothesize that, even in absence information, can occur as product surface-based grounded on orientation information derived from image cues such scale...

10.1167/5.8.341 article EN cc-by-nc-nd Journal of Vision 2010-03-16

Heitger et al (1998) proposed a neural model of illusory contour formation based on inputs from junctions and real edge orientations. A number recent results suggest common mechanism for at least some aspects occluded formation. These include the existence hybrid (quasimodal) contours cases in which interpolation can be shown to precede determination modal or amodal appearance (e.g., Petter effect). Moreover, ecological considerations that coping with occlusion (amodal completion) is likely...

10.1167/4.8.791 article EN cc-by-nc-nd Journal of Vision 2004-08-01
Coming Soon ...