Ph.D. candidate Email: topraj [at] gatech [dot] edu Office: TSRB 229 B I am currently a Ph.D. student in the Graphics group in the College of Computing at the Georgia Institute of Technology. Within the Graphics group, I am a part of the MAGIC lab. The Graphics group is in the School of Interactive Computing. I joined the PhD program at Georgia Tech in Fall 2007. I am co-advised by Professor Jarek Rossignac and Professor J. David Frost. Dr Frost heads the Geomaterial Surface and Structure Characterization Laboratory in the Geosystems Engineering group which is a part of the School of Civil and Environmental Engineering.

Publications:

	LR: Compact Connectivity Representation for Triangle Meshes Topraj Gurung, Mark Luffel, Peter Lindstrom, Jarek Rossignac Siggraph 2011 pdf | project page
	SQuad: Compact Representation for Triangle Meshes Topraj Gurung, Daniel Laney, Peter Lindstrom, Jarek Rossignac Eurographics 2011 pdf | project page
	SOT: Compact Representation for Tetrahedral Meshes Topraj Gurung, Jarek Rossignac 2009 SIAM/ACM Joint Conference on Geometric & Physical Modeling (SPM) pdf | slides | project page

Technical Reports:

SOT: Compact Representation for Triangle and Tetrahedral Meshes Topraj Gurung, Jarek Rossignac GT Technical Report pdf

Internships:

Data Analysis Group at Lawrence Livermore National Laboratory. Summer 2010. Worked together with Peter Lindstrom and Danel Laney on the SQuad data structure, spectral mesh processing applications and compact sparse matrix representations for meshes.

Projects on mesh generation, representation and compression:

Mesh generation

Hexahedralization [ongoing] We are exploring an innovative approach to hexahedralizing volumes. We approach it using spectral mesh processing techniques.

Mesh representation

	SOT: Compact Representation for Tetrahedral Meshes In SOT, we represent the connectivity of tetrahedral meshes using about 4 references per tetrahedron. We are able to achieve this reduction by matching a vertex to a tetrahedron and ordering the tetrahedra based on this matching.
	SOT: Compact Representation for Triangle Meshes In SOT, we represent the connectivity of triangle meshes using about 3 references per tetrahedron. We are able to achieve this reduction by matching a vertex to a triangle and ordering the triangles based on this matching.
	SQuad: Compact Representation for Triangle Meshes In SQuad, we represent the connectivity of meshes using about 2 references per triangle. We utilize the matching idea from the SOT and also pair triangles when possible. The quads allow us to infer connectivity information.

Mesh compression

	Tetrahedral Mesh Connectivity Compression We compress the connectivity of tetrahedral meshes by traversing the vertices in a breadth first manner while treating the star (tetrahedra incident on the vertex) of the vertex as a triangle mesh.
	Regular Triangle Mesh Compression Our goal is to compress the connectivity of triangle mesh where each vertex has valence 6. We assume the boundary also has valence 6 (assume that the mesh is part of a hole in the mesh). Given information about the boundary, we retrieve the interior (or the hole).

Projects on structural analysis:

	Segmentation of Particles and Pores We segment a 3d scanned voxel image of sand particles. The result of the segmentation is that we identify individual sand particles and individual pores. The particles can be approximated as ellipsoids.
	Analysis of geofibers [ongoing] Given a 3d scanned voxel image of fibrous material, we identify individual fibers and analyze the connectivity of the fibers.