Toshiya Hachisuka Assistant Professor toshiya_at_cs_au_dk Turing-123 Aabogade 34, DK-8200 Aarhus N, Denmark Pronunciation of my name: Almost literal, but "Toshiya" has its accent on "To" and "Hachisuka" has silent "u" and its accent is on "chi". I am looking for a PhD student who has research interests that well align with mine. International students are encouraged to apply.

• April/04/2012: I have posted a new technical report, "A Path Space Extension for Robust Light Transport Simulation". This paper basically combines BPT and SPPM using MIS.

• Q0. Your code for MDAS does not work with my pbrt.
• A0. Please read readme.txt again and you will get why.
  
  
• Q1. Is (S)PPM unbiased?
• A1. In the standard definition, (S)PPM is biased and consistent.
  
  
• Q2. Do you host summer interns?
• A2. If you have your own funding and be serious about it, then maybe.
  
  
• Q3. Do you speak Danish?
• A3. Not yet.

• Light transport simulation
• Physically based rendering
• Parallel processing
• Real-time rendering
• Physically based animation
• Computational statistics

• Tommy de Blanck (MS)
• Jacob Toft Pedersen (BS)
• Lasse Staal (MS)
• Horea Ioan Trinca (PhD - visiting)

• Rendering Algorithms (Q4, 2012)

	"Robust Adaptive Photon Tracing using Photon Path Visibility" T. Hachisuka and H. W. Jensen ACM Transactions on Graphics (Volume 30 Issue 5, expected to be presented at SIGGRAPH 2012 (postponed)), 2011 (Adaptive Markov chain Monte Carlo sampling for photon tracing with a simple target distribution.)
	"A Progressive Error Estimation Framework for Photon Density Estimation" T. Hachisuka, W. Jarosz, and H. W. Jensen ACM Transactions on Graphics (SIGGRAPH Asia 2010), 2010 (Full paper of the talk at SIGGRAPH 2010.)
	"Parallel Progressive Photon Mapping on GPUs" [code] [slides] T. Hachisuka and H. W. Jensen SIGGRAPH Asia 2010 Technical Sketches, 2010 (Simple parallel range query algorithm with randomization.)
	"An Error Estimation Framework for Photon Density Estimation" [slides] T. Hachisuka, W. Jarosz, and H. W. Jensen SIGGRAPH 2010 Talks: Rendering Intangibles, 2010 (Error estimation for progressive photon density estimation.)
	"Stochastic Progressive Photon Mapping" [code] [data0] [data1] T. Hachisuka and H. W. Jensen ACM Transactions on Graphics (SIGGRAPH Asia 2009), 2009 (First progressive average density estimator for distribution effects. )
	"An Empirical BSSRDF Model" C. Donner, J. Lawrence, R. Ramamoorthi, T. Hachisuka, H. W. Jensen, and S. Nayar ACM Transactions on Graphics (SIGGRAPH 2009), 2009 (Precomputed BSSRDF for moderately scattering media.)
	"Progressive Photon Mapping" [code] [data0] [data1] T. Hachisuka, S. Ogaki, and H. W. Jensen ACM Transactions on Graphics (SIGGRAPH Asia 2008), 2008 (New progressive density estimator for light transport simulation.)
	"Multidimensional Adaptive Sampling and Reconstruction for Ray Tracing" [code] [data0] [data1] T. Hachisuka, W. Jarosz, R. P. Weistroffer, K. Dale, G. Humphreys, M. Zwicker, and H. W. Jensen ACM Transactions on Graphics (SIGGRAPH 2008), 2008 (Adaptive sampling and reconstruction in the random number space.)
	"Combined Lagrangian-Eulerian Approach for Accurate Advection" [poster] [code] T. Hachisuka SIGGRAPH 2005 Posters (2nd Place in SRC:Undergraduate division at SIGGRAPH), 2005 (Advection equation solver using mapping functions.)
	"High-Quality Global Illumination Rendering Using Rasterization" [demo] T. Hachisuka GPU Gems 2: Programming Techniques for High Performance Graphics and General-Purpose Computation (Chapter 38), 2005 (Applicaton of the poster at SIGGRAPH 2004.)
	"Final Gathering on GPU" [poster] T. Hachisuka GP2, ACM Workshop on General Purpose Computing on Graphics Processors (also presented at SIGGRAPH 2004 Posters), 2004 (Complete GPU mapping of ray-bundle tracing.)

	"A Path Space Extension for Robust Light Transport Simulation" T. Hachisuka, J. Pantaleoni, and H. W. Jensen NVIDIA Technical Report NVR-2012-001, April 2012 (Unified sampling framework for Monte Carlo path integration and photon density estimation.)
	"Ray Tracing on Graphics Hardware" T. Hachisuka Research Exam, University of California at San Diego, 2009 (Comparative studies of GPU ray tracing algorithms.)
	"Generalized Polarization Ray Tracing using a Monte-Carlo Method" T. Hachisuka Final project in CSE 272, University of California at San Diego, 2007 (Monte Carlo formulation of polarization effects.)

	"Robust Light Transport Simulation using Progressive Density Estimation" T. Hachisuka (Advisor: Prof. Henrik Wann Jensen) Dissertation for Doctor of Philosophy, University of California at San Diego, 2011 (Light transport simulation framework based on progressive density estimation. )
	"Advection Equation Solver using Mapping Functions" T. Hachisuka (Advisor: Prof. Seiichi Koshizuka) Thesis for Bachelor of Engineering, The University of Tokyo, 2006 (Formulation of the advection equation as mapping.)

• Doctor of Philosophy in Computer Science, University of California at San Diego, Summer 2011
• Bachelor of Engineering, The University of Tokyo, Spring 2006

• Assistant Professor, Aarhus University, Fall 2011 to Present
• Software Intern, Nvidia, Summer 2010
• Developer Intern, Weta Digital, Summer 2009
• Contract Developer, nTechnology, Summer 2007
• Contract Researcher, Prometech Software, Spring 2005 to Spring 2006
• Contract Developer, Raia, Summer 2005 to Fall 2008
• Developer, Quat Technology, Summer 2003 to Winter 2003

• NVIDIA Fellowship, 2010 (chosen as a recipient also in 2008)
• ATI/AMD Fellowship, 2008
• Powell Foundation Fellowship, 2007
• Dean award of the Department of Engineering (summa cum laude equivalent), The University of Tokyo, 2006
• Award for Excellence in Science, JASSO, 2006
• 2nd Place at SIGGRAPH Undergraduate Division in ACM Student Research Competition, 2005
• Super Creator, IPA, 2003
• Exploratory Software Project for Youth (the MITOH Youth Program), IPA, 2002

• Reviewer for:
  • SIGGRAPH
  • SIGGRAPH Asia
  • Eurographics Symposium on Rendering
  • ACM Transaction on Graphics
  • Computers & Graphics
  • Computer Graphics Forum
  • Eurographics
  • High Performance Graphics
  • IEEE Computer Graphics and Applications
  • IEEE Transactions on Visualization and Computer Graphics
• Committee member for:
  • Eurographics Symposium on Rendering 2012
• Projects for fun
• Parthenon renderer
• My awesome coolest totally-wicked strategy for producing many papers (modification of an anonymous post):
  1. Preparation - Discover some new application domain (e.g., rendering of sushi).
  2. Paper #1 - Introduce the problem. Model it as a ray tracing problem and render images using pbrt. Minimal implementation effort is needed.
  3. Paper #2 - Solve the problem with importance sampling. Because you know something about your application domain, you can find a good probability density function to reduce variance a bit compared to #1.
  4. Paper #3 - Importance sampling is not working well in difficult cases, even with your good probability density function. Now let's solve it with a Markov Chain Monte Carlo method such as the Metropolis-Hasting algorithm.
  5. Paper #4 - Speed is important. Implement the algorithm on a GPU.
  6. Paper #5 - #4 is still too slow. You've found an interactive application of the problem (e.g., sushi bar), so now you need to focus on speed rather than image quality. Rasterization.
  7. Paper #6 - You've analyzed your application domain, and found that 50-70% of all problem instances belong to a special class that can be rendered more efficiently (e.g., subsurface scattering of fish meat is dominated by single scattering). New paper.
  8. Paper #7 - It's time for an approximation algorithm (e.g., subsurface scattering by texture-space blurring).
  9. Now, repeat the process from the step 2 by slightly changing the topic (e.g., rendering of sashimi).
• I am a minimalist.
• I like bulleted lists.