A l d o   Z a n g


Computer Graphics Postdoctoral Fellow
at IMPA

About Me

Developer and enthusiast photographer


Aldo Zang is a postdoctoral fellow at IMPA. He obtained both his M.Sc. and Ph.D. in Mathematics from the Institute for Pure and Applied Mathematics (IMPA) under the supervision of Prof. Luiz Velho. He attended the National University of Misiones (UNaM) where he graduated with honors in 2003 with a Licentiate degree in Mathematics. His research interests includes Photorealistic Rendering, Lighting capture and rendering, Image Processing, Geometric Processing, and BRDFs modeling.


Contact information


Institutional postal address

Aldo Zang
IMPA - Instituto de Matemática Pura e Aplicada
Estrada Dona Castorina, 110
Rio de Janeiro, RJ
22460-320, Brasil


OFFICE: rooms 126 and 305


Portfolio

Some of my works


Research


ARLuxrender (2010 - 2012)

ARLuxrender is an extension of the open source software Luxrender, which allows to render scenes with mixed real and virtual objects directly, without post-processing. To render a scene with these features is only necessary to have a photograph of the real scene, an HDR environment map of the light in the place where the artifical objects will be inserted, and a photograph of a pattern used to determine the position of the camera.

Adaptive 4-8 (a48) Mesh Data Structure Library (2012)

The a48 library is a template-based C++ library to support general mesh representations mainly in the Computer Graphics research field. The library provides three principal components: efficient polygonal mesh representation; mesh manipulation via stellar operations; and adaptive multi-resolution structure with simplification and refinement strategies. I am a collaborator in this project.

Direct Meshes Remeshing (2012)

In this project we develop a software for apply direct mesh remeshing using stellar operators.

QtCalib (2011 - 2012)

QtCalib is a software for camera calibration that uses the Tsai's algorithm. The "QtCalib Project" was developed by the VISGRAF Laboratory with financial support of CNPq.

Hybrid Sampling Machine (2009 - 2010)

The H.S.M (Hybrid Sampling Machine) was my master degree project. The key idea consists into use a hybrid sampling technique to get light samples from environment maps in the latitude-longitude format to use them in a physically based rendering algorithm. The HSM can be used to sample the entire map and obtain a set of directional lights that approximates the illumination of the map, or make a "Pre-sampled" map to be passed to the rendering program. HSM supports two algorithms, MedianCut of Paul Debevec, and Fast Hierarchical Importance Sampling of Ostroumokhov et al. For the second algorithm it was also used a pairwise clusterization scheme to eliminate redundant samples.

Publications


2015


Panoramas RGB-D Omnidirecionais com Múltiplas Camadas e suas Aplicações.

A. Zang, and L. Velho
SIBGRAPI 2015, Work in Progress. Salvador, Brasil.

Abstract:

We introduce a novel omnidirectional RGB-D panorama with multiple layers (PML) concept. This new representation opens a new range of applications for interactive display area, such as stereo viewing, disparity handling and parallax effects. This representation also offers new solutions for image based rendering problems.

2013


Rendering synthetic objects into full panoramic scenes using light-depth maps.

A. Zang, D. Felinto, and L. Velho
GRAPP 2013, poster. Barcelona, España.

Abstract:

This photo realistic rendering solution address the insertion of computer generated elements in a captured panorama environment. This pipeline supports productions specially aiming at spherical displays (e.g., fulldomes). Full panoramas have been used in computer graphics for years, yet their common usage lays on environment lighting and reflection maps for conventional displays. With a keen eye in what may be the next trend in the filmmaking industry, we address the particularities of those productions, proposing a new representation of the space by storing the depth together with the light maps, in a full panoramic light-depth map. Another novelty in our rendering pipeline is the one-pass solution to solve the blending of real and synthetic objects simultaneously without the need of post processing effects.

2012


Augmented Reality Using Full Panoramic Captured Scene Light-Depth Maps

A. Zang, D. Felinto, and L. Velho
SIGGRAPH Asia 2012, poster. Singapore.

Abstract:

Augmented reality rendering with the reconstructed depth of the scene incorporated in a light-depth map for more accurate shadows and reflections of synthetic elements into a full panoramic scene.


Production framework for full panoramic scenes with photo-realistic augmented reality

D. Felinto, A. Zang, and L. Velho
CLEI 2012. Medellin, Colombia.

Abstract:

The novelty of our workflow is the end-to-end solution to combine captured panoramas and computer generated elements. This pipeline supports productions specially aiming at spherical displays (e.g., fulldomes). Full panoramas have been used in computer graphics for years, yet their common usage lays on environment lighting and reflection maps for conventional displays. With a keen eye in what may be the next trend in the filmmaking industry, we address the particularities of those productions, proposing a new representation of the space by storing the depth together with the light maps, in a full panoramic light-depth map.

2011


Um framework para renderizações foto-realistas de cenas com realidade aumentada

A. Zang and L. Velho
CLEI 2011. Quito, Equador.

Abstract:

Scenes with augmented reality are used daily in all kinds of applications ranging from film, television, advertising, up scientific simulations. The main problem of building an photo-realistic augmented scene is to achieve a photo realistic result through a single rendering, without need post-production. Aiming this problem, we propose a rendering framework for augmented scenes where the final scene can be calculated in a single direct rendering without need post processing. The main novelty in this framework is the way to perform rendering real and synthetic objects simultaneously.

2010


Techniques for CG music video production: the making of Dance to the Music / Play to the Motion

A. Schulz, M. Cicconet, B. Madeira, A. Zang, and L. Velho
Visgraf Technical Report 2010. Rio de Janeiro, Brasil

Abstract:

This technical report presents a set of techniques for the production of music videos using computer graphics and animation. Their practical application is illustrated through examples developed for making the video “Dance to the Music / Play to the Motion”.


Esquema híbrido para amostragem de mapas de iluminação em renderizações foto-realistas

A. Zang and L. Velho
First Prize at SIBIGRAPI-WTD 2009. Rio de Janeiro, Brasil.
Second Prize at CLEI-CLTM 2010. Assunción, Paraguay.

Abstract:

In this paper we introduce a new approach to the problem of direct illumination in physically-based rendering of 3D scenes using illumination maps captured from real environments. We developed a system that takes advantage of the best features of the current solutions to the problem: namely, the approximation of illumination maps through directional lights; and stochastic sampling of the light maps. Our framework is flexible and can be used with most rendering programs.

2009


Esquema híbrido para amostragem de mapas de iluminação em renderizações foto-realistas

A. Zang - (Master thesis, 2009)
Advisor: L. Velho

Abstract:

Currently, photorealistic rendering of synthetic 3D scenes using illumination maps captured from real environments is an important topic in image synthesis. There is also a great demand for efficient solutions, these techniques are used for incorporation of synthetic objects in films and commercial.
In this thesis we introduce a new approach to the problem of direct illumination in physically based rendering of 3D scenes using illumination maps captured from real environments. We developed a system that takes advantage of the best features of the current solutions to the problem: namely, the approximation of illumination maps through directional lights; and stochastic sampling of the light maps. Our framework is flexible and can be used with most rendering programs.
The proposed Hybrid Scheme can be used with both direct illumination methods and algorithms for global illumination. Our proposal will complement and consolidate the traditional techniques, because they can be used either separately or combined to address the problem of sampling light sources during rendering with real lighting.

2008


Sistema composto para amostragem e geração de luzes a partir de mapas de iluminação

A. Zang and L. Velho
SIBIGRAPI-WTDCGPI, 2008. Campo Grande, MS, Brasil.

Abstract:

In this paper we introduce a new approach to the problem of direct illumination in physically-based rendering of 3D scenes using illumination maps captured from real environments. We developed a system that takes advantage of the best features of the current solutions to the problem: namely, the approximation of illumination maps through directional lights; and stochastic sampling of the light maps. Our framework is flexible and can be used with most rendering programs.

Courses Projects and other things


Image Processing (2012-2)

In imaging science, image processing is any form of signal processing for which the input is an image, such as a photograph or video frame; the output of image processing may be either an image or a set of characteristics or parameters related to the image. Most image-processing techniques involve treating the image as a two-dimensional signal and applying standard signal-processing techniques to it.

Geometry Processing (2012-1)

Geometry processing, or mesh processing, is a fast-growing area of research that uses concepts from applied mathematics, computer science and engineering to design efficient algorithms for the acquisition, reconstruction, analysis, manipulation, simulation and transmission of complex 3D models. Applications of geometry processing algorithms already cover a wide range of areas from multimedia, entertainment and classical computer-aided design, to biomedical computing, reverse engineering and scientific computing.

Computational Manifolds and Applications (2011-2)

This program aims at the study of two-dimensional manifolds and applications to geometric modeling and visualization. The course is an initiative of IMPA inspired in the traditional scientific programs that are organized by the MSRI, and the Schloss Dagstuhl.

Computational Methods in Optimization (2010-2)

Problem: find the provision that should have $k$ points on the unit sphere in order to minimize the potential energy. The potential energy of a set $\{x_1, x_2 , \ldots, x_k \}$ of $k$ points on the unit sphere is computed as $$\sum_{1 \leq i \lt j \leq k} \frac{1}{||x_i - x_j||}.$$

Computer Graphics Seminar (2007-2)

Presentation of paper "Fast Hierarchical Importance Sampling whit Blue Noise Properties" of Ostromoukhov et al. in the Computer Graphics Seminar at IMPA.

Computational Geometry (2007-1)

The purpose of the Computational Geometry is study geometric problems under the algorithmic standpoint. There is a parallel between Computational Geometry and Geometric Design: on both the objective often is to obtain new geometric elements from elementary buildings. The difference is that in Computational Geometry, geometrical figures and constructions represent data structures and algorithms, respectively. The problem of interest, in general, is to perform a particular construction to use the smallest possible number of elementary steps.

3D Graphics Systems (2007-1)

This course is an introductory graphics course emphasizing the practical aspects of the discipline. It covers the fundamentals of modeling and rendering of surfaces in three dimensions. These topics are presented in the context of a computer graphics system.

Image Processing (2006-2)

The Course is part of the graduate program on Computer Graphics maintained by IMPA and PUC-Rio. It is a regular course for the master program, and it is also recommendend for first year graduate students of the program.