Google Scholar [www]


Chronological order: 3DOR'19,GMOD'19,TVCJ'19,JRTIP'19,
TOG'18 (SIGGRAPH Asia), TVCJ'18a, TVCJ'18b, GMOD'18,
CGA'17, TJEECS'17,
TOG'16 (SIGGRAPH), 3DOR'16a, 3DOR'16b, MM'16a, MM'16b,
MIA'15, C&G'15, MM'15,
CGF'14a (Pacific Graphics), CGF'14b,
CGF'13,
PAMI'12, CGF'12 (Pacific Graphics), PhD'12,
CGF'11 (Symposium on Geometry Processing),
CVPR'10, CVIU'10, CGVR'10, SIU'10,
PRL'09, CGVR'09,
MSc'06.


3D Correspondence

Recent Advances in Shape Correspondence

Y. Sahillioğlu
The Visual Computer, 2019

[pdf]
Input n/a
Output n/a
Idea Wrapping up all the interesting and brand new shape correspondence methods of the 2011-2019 interval.

A Genetic Isometric Shape Correspondence Algorithm with Adaptive Sampling

Y. Sahillioğlu
ACM Transactions on Graphics, (Presented at SIGGRAPH Asia), Vol. 37, No. 5, 175, 2018

[pdf]  [ppt *]  [code and executable]  [video1]  [video2]
Input Isometric (or nearly isometric) shape pairs
Output Coarse correspondence b/w samples on two shapes
Idea Exploring the space of permutations wisely through a genetic algorithm; improve a given map further with the adaptive sampling scheme

SHREC’19: Shape Correspondence with Isometric and Non-Isometric Deformations

R. Dyke, C. Stride, .., Y. Sahillioğlu, .., J. Yang
In Proc. of Eurographics Workshop on 3D Object Retrieval (3DOR), 2019

[pdf]  [ppt *]  [code and executable]  [video1]  [video2]
Input Isometric (or nearly isometric) shape pairs
Output Coarse correspondence b/w samples on two shapes
Idea Putting my TOG'18 algorithm to a test on SHREC contest: If pairs with geodesic inconsistencies and symmetric flips were excluded, my results would have improved significantly (see the penultimate paragraph of Section 4)

SHREC'16: Matching of Deformable Shapes with Topological Noise

Z. Lahner, E. Rodola, M. M. Bronstein, D. Cremers, O. Burghard, L. Cosmo, A. Dieckmann, R. Klein, Y. Sahillioğlu
In Proc. of Eurographics Workshop on 3D Object Retrieval (3DOR), 2016

[pdf]  [code and executable
Input Isometric (or nearly isometric) shape pairs with topological noise
Output Coarse correspondence b/w evenly-spaced high-curvature samples on two shapes
Idea Minimizing isometric distortion in probabilistic EM framework using topologically-robust biharmonic distance

SHREC'16: Partial Matching of Deformable Shapes

L. Cosmo, E. Rodola, M. M. Bronstein, A. Torsello, D. Cremers, Y. Sahillioğlu
In Proc. of Eurographics Workshop on 3D Object Retrieval (3DOR), 2016

[pdf]  [code and executable
Input Partially isometric shape pairs where the partial model is an isometric subset of the complete one
Output Partial correspondence b/w two shapes
Idea Introducing a distortion measure for partial matching based on raw geodesic distances

Multiple Shape Correspondence by Dynamic Programming

Yusuf Sahillioğlu and Yücel Yemez
Computer Graphics Forum (Proc. PG), Vol. 33, No. 7, 121-130, 2014

[pdf]  [ppt]  [code and executable]  [video]
Input A collection of isometric (or nearly isometric) shapes
Output Set of consistent maps b/w all shape pairs w/ the least overall isometric distortion
Idea Finding optimal path of nodes on a graph where each node is a pairwise map

Partial 3D Correspondence from Shape Extremities

Yusuf Sahillioğlu and Yücel Yemez
Computer Graphics Forum, Vol. 33, No. 6, 63-76, 2014

[pdf]  [code and executable]
Input Isometric (or nearly isometric or partially isometric) shape pairs; partial shapes w/ uncommon parts are welcome
Output Partial or complete correspondences b/w two shapes
Idea Making triplets of qualified matches produce votes based for matching in the presence of uncommon parts

Coarse-to-Fine Isometric Shape Correspondence by Tracking Symmetric Flips

Yusuf Sahillioğlu and Yücel Yemez
Computer Graphics Forum, Vol. 32, No. 1, 177-189, 2013

[pdf]  [code and executable]
Input Isometric (or nearly isometric) shape pairs
Output Correspondence b/w two shapes at desired resolution
Idea Handling the symmetric flip problem by tracking multiple maps upto a level of detail

Scale Normalization for Isometric Shape Matching

Yusuf Sahillioğlu and Yücel Yemez
Computer Graphics Forum (Proc. PG), Vol. 31, No. 7, 2233-2240, 2012

[pdf]  [ppt]  [code and executable]
Input Isometric (or nearly isometric or partially isometric) shape pairs; partial model is an isometric subset of the complete one
Output Partial or complete correspondences b/w two shapes
Idea Introducing a distortion measure for partial matching based on unnormalized geodesic distances; use the fact that ratios between geodesic distances are preserved under scaling and isometric deformations

Minimum-Distortion Isometric Shape Correspondence Using EM Algorithm

Yusuf Sahillioğlu and Yücel Yemez
IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI), Vol. 34, No. 11, 2203-2215, 2012

[pdf]  [code and executable]
Input Isometric (or nearly isometric) shape pairs
Output Coarse correspondence b/w evenly-spaced high-curvature samples on two shapes
Idea Minimizing isometric distortion in probabilistic EM framework

Coarse-to-Fine Combinatorial Matching For Dense Isometric Shape Correspondence

Yusuf Sahillioğlu and Yücel Yemez
Computer Graphics Forum (Proc. SGP), Vol. 30, No. 5, 1461-1470, 2011

[pdf]  [ppt]  [code and executable]
Input Isometric (or nearly isometric) shape pairs
Output Correspondence b/w two shapes at desired resolution
Idea Subdividing matched patches recursively into smaller patches to be matched

3D Shape Correspondence by Isometry-Driven Greedy Optimization

Yusuf Sahillioğlu and Yücel Yemez
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 453-458, 2010

[pdf]  [code and executable]  [video]
Input Isometric (or nearly isometric) shape pairs
Output Coarse correspondence b/w evenly-spaced samples on two shapes
Idea Exposing the initial spectral correspondence to a greedy refinement

Algorithms for 3D Isometric Shape Correspondence

Yusuf Sahillioğlu
PhD Thesis, Computer Science Dept., Koç University, Turkey, August 2012
(Graduate studies excellence award)

[pdf]  [ppt *
Input Isometric (or nearly isometric or partially isometric) shape pairs
Output Complete or partial correspondences according to the given scenario
Idea PhD

3D Deep Learning

Deep 3D semantic scene extrapolation

Ali Abbasi and Sinan Kalkan and Yusuf Sahillioğlu
The Visual Computer, Vol. 35, No. 2, 271-279, 2019 (online: 2018)

[pdf]  [code and executable]
Input Left half of a 3D scene
Output Extrapolated right half of the scene
Idea Learning extrapolation from a deep net with a convenient loss function

3D Deformation

Detail-preserving Mesh Unfolding for Nonrigid Shape Retrieval

Yusuf Sahillioğlu and Ladislav Kavan
ACM Transactions on Graphics, (Presented at SIGGRAPH), Vol. 35, No. 3, 27, 2016

[pdf]  [ppt *]  [code and executable]  [supplementary material]  [video]
Input A tetrahedral mesh to be unfolded with preserved details
Output Tetrahedral and triangular meshes in the detail-preserved unfolded pose
Idea Using mass-spring system to separate each mesh vertex as far as possible while preserving original edge lengths; additional finite element constraints enhace the regularization

Skuller: A Volumetric Shape Registration Algorithm for Modeling Skull Deformities

Yusuf Sahillioğlu and Ladislav Kavan
Medical Image Analysis, Vol. 23, No. 1, 15-27, 2015

[pdf]  [ppt]  [code and executable]  [video]
Input Subject-specific CT scan data and a generic skull model
Output Deformed skull model conformed to the CT data
Idea Using volumetric models on both sides of the registration pipeline

A Shape Deformation Algorithm for Constrained Multidimensional Scaling

Yusuf Sahillioğlu
Computers & Graphics, Vol. 53, 156-165, 2015

[pdf]  [code and executable]  [video]
Input A tetrahedral mesh to be brought into its MDS pose with preserved details
Output Tetrahedral and triangular meshes in the detail-preserved MDS pose
Idea Guiding the deformation by MDS representation using a fast sparse linear system

3D Reconstruction

Coarse-to-Fine Surface Reconstruction from Silhouettes and Range Data Using Mesh Deformation

Yusuf Sahillioğlu and Yücel Yemez
Computer Vision and Image Understanding (CVIU), Vol. 114, 334-348, 2010

[pdf
Input Real-world object along w/ its multi-view silhouettes and range data
Output Visual hull and cavity-sensitive refined mesh of the object
Idea Moving the intersected triangle from the visual hull surface towards the range surface

Triangulation-free 3D Reconstruction from LiDAR Data

Yusuf Sahillioğlu
International Conference on Computer Graphics & Virtual Reality (CGVR), 27-32, 2010

[pdf]
Input Unorganized, noisy, and dense 3D points acquired by a LiDAR system
Output Low-resolution 2-manifold triangular mesh approximating the LiDAR surface
Idea Guiding the deformation w.r.t. the best-fit tangent planes spread over LiDAR points

Shape from Silhouette Using Topology-Adaptive Mesh Deformation

Yücel Yemez and Yusuf Sahillioğlu
Pattern Recognition Letters, Vol. 30, 1198-1207, 2009

[pdf
Input Real-world object along w/ its multi-view silhouettes and range data
Output Visual hull of the object
Idea Projecting each 3D vertex of the deforming surface to the silhouettes in order to guide the topology-adaptive deformation

A Surface Deformation Framework for 3D Shape Recovery

Yusuf Sahillioğlu
MS Thesis, Computer Science Dept., Koç University, Turkey, 2006

[pdf]   [ppt]
Input Multi-view silhouettes and/or range data of a real-world object to be reconstructed
Output Reconstruction, i.e., visual hull and/or its refined version, as a 2-manifold triangular mesh
Idea Deforming the bounding sphere in the guidance of silhouettes and then refining it further w/ scan lines of the range data

3D Rendering

Voxel transformation: scalable scene geometry discretization for global illumination

B. Yalçıner and Y. Sahillioğlu
Journal of Real-Time Image Processing, 2019

[pdf]  [video]
Input A dynamic scene with many animated objects
Output Realistic and fast rendering of the scene
Idea Transforming pre-generated voxel data from model space to world space, which is in contrast to the common (and slower) way of voxelizing each dynamic object over each frame

3D Retrieval

3D indirect shape retrieval based on hand interaction

Erdem Can Irmak and Yusuf Sahillioğlu
The Visual Computer, 2018

[pdf]
Input Hand pose captured by the Leap Motion device and a 3D model database to explore
Output Set of 3D shapes that are similar to the query hand pose
Idea Learning the optimal parameters that encode the way humans grab certain objects and using them for interaction-based shape retrieval

An Evaluation of Canonical Forms for Non-Rigid 3D Shape Retrieval

D. Pickup, J. Liu, X. Sun, P. Rosin, R. Martin, Z. Cheng, Z. Lian, S. Nie, L. Jin, G. Shami, Y. Sahillioğlu, L. Kavan
Graphical Models, Vol. 97, 17-29, 2018

[pdf]  [code and executable]
Input Pose-independent surface or volume mesh of a 3D shape to be retrieved and a 3D model database to explore
Output Set of 3D shapes that are similar to the query shape up to articulations
Idea Evaluating my existing algorithms (C&G'15 and TOG'16) in a comprehensive test suite

Sketch-based Articulated 3D Shape Retrieval

Yusuf Sahillioğlu and Metin Sezgin
IEEE Computer Graphics and Applications, Vol. 37, No. 6, 88-101, 2017

[pdf]  [code and executable]
Input Pose-independent 2D sketch of a 3D shape to be retrieved and a 3D model database to explore
Output Set of 3D shapes that are similar to the query sketch up to articulations
Idea Using our good continuation rule on sketches to facilitate articulation-invariant comparisons

3D Skeleton

3D Skeleton Transfer for Meshes and Clouds

Ç. Seylan and Y. Sahillioğlu
Graphical Models, Vol. 105, 2019

[pdf]  [supplementary material]  [data]
Input Source mesh and its skeleton, and the target mesh or point cloud without skeleton
Output 1D skeleton inside the target mesh or point cloud
Idea Transferring the source skeleton to the target model, which can be a watertight or punctured surface mesh, or a cloud.

Others

My invited Eurasia Graphics 2018 Workshop talk on 3D Printing: [pdf] [ppt]
My invited TOBB Computer Engineering Department seminar talk on Shape Matching: [pdf] [ppt]
My instructive presentation on Range Queries and Square Root Complexity: [pdf] [ppt]
My instructive presentation on Machine Learning and Graph Coloring: [pdf]
My C++ implementation of the ICP algorithm: [code and executable]
A marching algorithm for isosurface extraction from face-centered cubic lattices, Y. Sahillioğlu, Turkish Journal of Electrical Engineering and Computer Sciences (TJEECS), 2017. [pdf] [code and executable]
iAutoMotion - an Autonomous Content-based Video Retrieval Engine, L. Rossetto, I. Giangreco, C. Tanase, H. Schuldt, O. Seddati, S. Dupont. T. M. Sezgin, Y. Sahillioğlu, 22nd International Conference on Multimedia Modeling, 2016. [pdf]
IMOTION - Searching for Video Sequences using Multi-Shot Sketch Queries, L. Rossetto, I. Giangreco, S. Heller, C. Tanase, H. Schuldt, O. Seddati, S. Dupont. T. M. Sezgin, O. C. Altıok, Y. Sahillioğlu, 22nd International Conference on Multimedia Modeling, 2016. [pdf]
IMOTION: A Content-based Video Retrieval Engine, L. Rossetto, I. Giangreco, H. Schuldt, S. Dupont, O. Seddati, T. M. Sezgin, Y. Sahillioğlu, 21st International Conference on Multimedia Modeling, 2015. [pdf]
3B İzometrik Şekil Eşleme, Yusuf Sahillioğlu and Yücel Yemez, IEEE Sinyal İşleme ve Uygulamaları Kurultayı (SİU), 2010 [pdf] (Best student paper)
3D Correspondence by Breadth-First Search Frontiers, Yusuf Sahillioğlu, Int. Conference on Computer Graphics & Virtual Reality (CGVR), 203-207, 2009 [pdf]
A Surface Deformation Framework for 3D Shape Recovery, Yusuf Sahillioğlu and Yücel Yemez, Lecture notes in Computer Science (MCRS), 4105, 570-577, 2006 [pdf]
Çok Kameralı Video Görüntülerinden Yüzey Deformasyonu ile 3B Şekil Geri Çatma ve İzleme, Yusuf Sahillioğlu and Yücel Yemez, SİU, 2006 [pdf]
Hair Motion Simulation, Yusuf Sahillioğlu and Bülent Özgüç, Int. Symp. on Computer and Inf. Sciences (ISCIS), 126-135, 2004 [pdf]


Back to home page.