Confessions of a researchaholic

2011-11-20

Rejection

Filed under: Real — liyiwei @ 7:36 pm
Tags: , , ,

Let me begin by outlining a few basic laws of nature.

Scarcity
Resources are scarce, with scarcity proportional to relative supply and demand.

Ego
People tend to consider themselves better than they really are (and thus deserve better than what they currently have).

Randomness
The universe is stochastic, and unfortunately people like to imagine life is more deterministic than it really is. (Heck, even Einstein refused to believe God will throw dice.)

As a consequence, when people cannot get what they want, they complain about injustice, and question the fairness of the process that allocates the scarce resources.

Unfortunately, none of these will not help or change anything. People will just think you are a whiner. And do not even attempt to ask why you did not get what you want; you are forcing people to either tell you a white lie (e.g. you are really good; it is just that we cannot figure out a good task for you to do right now) or be impolite (e.g. dude, you really have no idea how lousy your performance was?) or to confess they really have no clue (e.g. huh, we cannot really decide between you and another candidate, so we picked the taller one).

Specifically, under a lot of situations, the decision depends on a lot of factors. For example, there is this famous thing called *affirmative action* in the US that discourages discrimination in job and school opportunities. The intention is all good, until people start to enforce hard quotas, like reserving a certain percentage of openings to a certain group of (alleged) minority. Then, if the real application pool does not contain that percentage of qualified minorities, the process will deem to hire a minority with inferior ability than certain more qualified majority. You can say it is not fair, but that is how things work.
Sometimes it can just be random; with hundreds or even thousands of applications for a certain opening, it is just information overload to for anyone to decide who is really good.

My personal suggestion is to stop worrying all of these, and use the rejection or denial as a source of motivation. Prove to the world that they are wrong, and you are the best.
Getting dumped by your ex? Shine yourself to make him/her regret next time you meet.
Getting rejected by a job application? Outperform whoever got hired instead of you.
Did not get permission to your dream school or professor? Publish better research papers than his/her group.

These being said, I really want to tell whoever applied to be my student: thank you all very much, but since I have received hundreds of applications and I can take probably only 2 (or maybe one more if I can muscle with the department), mathematically it is impossible for me to take everyone I like. Evaluating the quality of a student is more art than science, so I rely mostly on my intuition. If you really want to force an answer out of me, the only thing I can say is that my intuition told me that there are better and more suitable candidates than you. But can I prove it? Heck no, and I doubt if anyone can.

Ray tracing

Filed under: Real — liyiwei @ 6:43 pm
Tags: ,

Many mentors have their own favorite entry level project for both training and sieving purposes. And mine is ray tracing.

Read + code

Read the following books and implement your own ray tracer from scratch.
Yes, I mean writing every single line of the code yourself, except for standard libraries like iostream and cmath. And yes, I recommend c++ as the programming language.

Ray tracing in one weekend, by Peter Shirley.
Path tracing with soft phenomena. Very short and cute introduction to ray tracing in particular and computer graphics in general. Full of hidden gems and wisdom, but beginners might not appreciate all the nuances.

An introduction to ray tracing, by Eric Haines, Pat Hanrahan, Rob Cook, Jim Arvo, David Kirk, Paul Heckbert, and (edited by) Andrew Glassner.
Whitted-style ray tracing with hard phenomena. More comprehensive than the book above.

You should read the books (especially the first one) on a computer and code along. Do not just read; it is less fun and less effective.
Software engineer your code from the very beginning; ray tracing is nicely object oriented with natural object-like components such as rays, shapes, materials, colors, motions, and cameras.

Demo

Produce your own ray traced images. Start with something simple, like mirrors and glasses for basic reflections and refractions. Ultimately, I want to see 2 things:

  • Produce ray traced animations from the BART benchmark. (If you can reach this stage, I consider your training complete. If you cannot even get to this point for whatever reason, you have no chance to survive in our future projects.)
  • Design your own favorite scene, and ray trace it. Show me your creativity and artistic + scientific blend. You can find examples from prior Stanford cs348b rendering competitions.
    (This part can be good for a group project, even though I expect everyone to complete everything else individually.)

Why

I did ray tracing as part of Stanford CS348b image synthesis, and I would say this is perhaps the class that I have learned the most from.
First, ray tracing is a superb training for coding and software engineering; it is inherently modular and suitable for object oriented programming, and the amount of coding is non-trivial (actually, quite hefty for new-comers, especially if you code from scratch).
Second, ray tracing is a superb introduction to image formation, and you will learn the fundamentals of physics, computer graphics, computer vision, and image processing.
Third, and perhaps most importantly, it is heck of fun. It is beyond words for me to describe the satisfaction I (and many people I heard from) derived from producing my first ray traced image. I am not an artist, but I can paint with algorithms.

P.S.

There is another excellent book for ray tracing: Physically based rendering, by Matt Pharr and Greg Humphreys, but I personal opinion is that you should read it as your second, not first, ray tracing book.

2011-10-12

Why I do not stay up late for work

Filed under: Real — liyiwei @ 9:07 am
Tags: , , ,

Once, while in grad school, I tried to wrap up a long SIGGRAPH crunching day by cleaning up my code.

So, I did the following in the Makefile:

clean:
rm -r -f *.c

Instead of

clean:
rm -r -f *.o

After that wonderful experience, I decided not to stay up late for work anymore. 🙂

2011-09-17

Producing research results

Filed under: Real — liyiwei @ 4:06 pm
Tags: ,

Sylvain Lefebvre has a nice post about how to structure code for producing research results; see here. (Full disclosure: Sylvain is my collaborator, but he is a superb researcher and coder, as evident from his resume.)

2011-05-16

SIGGRAH contributor discount

Filed under: Real — liyiwei @ 3:18 pm
Tags:

SIGGRAPH gives one contributor code to each technical paper so that at most one author can get a 50 percent registration discount. However, the discount is not accumulative, so (in effect) each author can receive at most one contributor code. By considering the fact that some authors may be on multiple papers, it becomes an interesting optimization problem on how to assign the contributor codes among all paper authors to maximize the total number of utilized discounts.

Under the general scenarios, this looks like an NP hard problem. (Correct me if I am wrong, as I am very rusty about basic algorithm theories.) By constructing a pathological scenario with the right number of total authors and authors on each paper, this can be quite a difficult problem to solve.

Fortunately, the reality is much simpler; since the number of papers for each author is usually quite limited, the *relationship graph* among all authors contain many isolated islands (that can be solved independently). In fact, I have found that a very simple randomized algorithm would just work: visit the papers in a random order, and pick a random author that has yet received any contributor code.

The results for SIGGRAPH 2011 technical papers are as follows. For each paper, blue color indicates the author receiving the contributor code, red color indicates authors who receive codes from other papers, and gray color indicates authors who receive nothing.

(I hope ACM won’t ban me from attending SIGGRAPH.)

Physically-valid Statistical Motion Models for Human Motion Synthesis
Xiaolin Wei
Jianyuan Min
Jin-Xiang Chai

Motion Capture From Body-Mounted Cameras
Takaaki Shiratori
Hyun Soo Park
Leonid Sigal
Yaser Sheikh
Jessica Hodgins

Motion Reconstruction Using Sparse Accelerometer Data
Jochen Tautges
Arno Zinke
Bjorn Kruger
Jan Baumann
Andreas Weber
Thomas Helten
Meinard Muller
Hans-Peter Seidel
Bernd Eberhardt

Video-based Characters – Creating New Human Performances from a Multi-view Video Database
Feng Xu
Yebin Liu
Carsten Stoll
James Tompkin
Gaurav Bharaj
Qionghai Dai
Hans-Peter Seidel
Jan Kautz
Christian Theobalt

ShadowDraw: Real-Time User Guidance for Freehand Drawing
Yong Jae Lee
Larry Zitnick
Michael F. Cohen

OverCoat: An Implicit Canvas for 3D Painting
Johannes Schmid
Martin Sebastian Senn
Markus Gross
Robert Sumner

A Programmable System for Artistic Volumetric Lighting
Derek Nowrouzezahrai
Jared Johnson
Andrew Selle
Dylan Lacewell
Michael Kaschalk
Wojciech Jarosz

Coherent Noise for Non-Photorealistic Rendering
Michael Kass
Davide Pesare

Exploration of Continuous Variability in Collections of 3D Shapes
Maksim Ovsjanikov
Wilmot Li
Leonidas Guibas
Niloy J. Mitra

Characterizing Structural Relationships in Scenes Using Graph Kernels
Matthew Fisher
Manolis Savva
Pat Hanrahan

Probabilistic Reasoning for Assembly-Based 3D Modeling
Siddhartha Chaudhuri
Evangelos Kalogerakis
Leonidas Guibas
Vladlen Koltun

Shape google: Geometric words and expressions for invariant shape retrieval
Alexander Bronstein
Michael Bronstein
Leonidas Guibas
Maksim Ovsjanikov

Eulerian Solid Simulation With Contact
David I.W. Levin
Joshua Litven
Garrett L. Jones
Shinjiro Sueda
Dinesh Pai

Efficient Elasticity for Character Skinning with Contact and Collisions
Aleka McAdams
Yongning Zhu
Andrew Selle
Mark Empey
Rasmus Tamstorf
Joseph M. Teran
Eftychios Sifakis

Toward High-Quality Modal Contact Sound
Changxi Zheng
Doug James

A Nonsmooth Newton Solver for Capturing Exact Coulomb Friction in Fiber Assemblies
Florence Bertails-Descoubes
Florent Cadoux
Gilles Daviet
Vincent Acary

Large-Scale Dynamic Simulation of Highly Constrained Strands
Shinjiro Sueda
Garrett L. Jones
David I.W. Levin
Dinesh Pai

HDR-VDP-2: A calibrated visual metric for visibility and quality predictions in all luminance conditions
Rafal Mantiuk
Kil Joong Kim
Allan G. Rempel
Wolfgang Heidrich

A Versatile HDR Video Production System
Mike Tocci
Chris Kiser
Nora Tocci
Pradeep Sen

Perceptually Based Tone Mapping for Low-Light Conditions
Adam G. Kirk
James F. O’Brien

Illumination Decomposition for Material Recoloring with Consistent Interreflections
Rob Carroll
Ravi Ramamoorthi
Maneesh Agrawala

High-Quality Spatio-Temporal Rendering using Semi-Analytical Visibility
Carl Johan Gribel
Rasmus Barringer
Tomas Akenine-Moller

Frequency Analysis and Sheared Filtering for Shadow Light Fields of Complex Occluders
Kevin Egan
Florian Hecht
Fredo Durand
Ravi Ramamoorthi

Temporal Light Field Reconstruction for Rendering Distribution Effects
Jaakko Lehtinen
Timo Aila
Jiawen Chen
Samuli Laine
Fredo Durand

The Area Perspective Transform: A Homogeneous Transform for Efficient In-Volume Queries
Warren Hunt
Greg Johnson

Building Volumetric Appearance Models of Fabric using Micro CT Imaging
Shuang Zhao
Wenzel Jakob
Steve Marschner
Kavita Bala

Pocket Reflectometry
Peiran Ren
Jiaping Wang
John Snyder
Xin Tong
Baining Guo

Microgeometry Capture Using an Elastomeric Sensor
Micah Johnson
Forrester Cole
Alvin Raj
Edward Adelson

CATRA: Interactive Measuring and Modeling of Cataracts
Vitor F. Pamplona
Erick Baptista Passos
J Zizka
Manuel M. Oliveira
Esteban Clua
Ramesh Raskar

Blue-Noise Point Sampling using Kernel Density Model
Raanan Fattal

Efficient Maximal Poisson-Disk Sampling
Mohamed S. Ebeida
Anjul Patney
Scott A. Mitchell
Andrew Davidson
Patrick M. Knupp
John D. Owens

Differential Domain Analysis for Non-uniform Sampling
Li-Yi Wei
Rui Wang

Filtering Solid Gabor Noise
Ares Lagae
George Drettakis

GlobFit: Consistently Fitting Primitives by Discovering Global Relations
Yangyan Li
Xiaokun Wu
Yiorgos Chrysanthou
Andrei Sharf
Daniel Cohen-Or
Niloy J. Mitra

Global Registration of Dynamic Range Scans for Articulated Model Reconstruction
Will Chang
Matthias Zwicker

Texture-Lobes for Tree Modelling
Yotam Livny
Soeren Pirk
Zhanglin Cheng
Feilong Yan
Oliver Deussen
Daniel Cohen-Or
Baoquan Chen

L1-sparse Reconstruction of Sharp Point Set Surfaces
Haim Avron
Andrei Sharf
Chen Greif
Daniel Cohen-Or

Interactive and Anisotropic Geometry Processing Using the Screened Poisson Equation
Ming Chuang
Michael Kazhdan

Dinus: Double insertion, nonuniform, stationary subdivision surfaces
Kerstin Muller
Christoph Funfzig
Lars Reusche
Dianne Hansford
Gerald Farin
Hans Hagen

An Efficient Scheme for Curve and Surface Construction Based on a Set of Interpolatory Basis Functions
Ren-Jiang Zhang
Weiyin Ma

A Quantized-Diffusion Model for Rendering Translucent Materials
Eugene d’Eon
Geoffrey Irving

A Comprehensive Theory of Volumetric Radiance Estimation using Photon Points and Beams
Wojciech Jarosz
Derek Nowrouzezahrai
Iman Sadeghi
Henrik Wann Jensen

Progressive Photon Mapping: A Probabilistic Approach
Claude Knaus
Matthias Zwicker

Cache-Oblivious Ray Reordering
Bochang Moon
Youngyong Byun
Tae-Joon Kim
Pio Claudio
Hye-sun Kim
Yun-ji Ban
Seung Woo Nam
Sung-Eui Yoon

Expression Flow for 3D-Aware Face Component Transfer
Fei Yang
Jue Wang
Eli Shechtman
Lubomir Bourdev
Dimitris Metaxas

Image-Guided Weathering: A New Approach Applied to Flow Phenomena
Carles Bosch
Pierre-Yves Laffont
Holly Rushmeier
Julie Dorsey
George Drettakis

Exploring Photobios
Ira Kemelmacher-Shlizerman
Rahul Garg
Steven M. Seitz
Eli Shechtman

Discrete Element Textures
Chongyang Ma
Li-Yi Wei
Xin Tong

Space-time Planning with Parameterized Locomotion Controllers
Sergey Levine
Yongjoon Lee
Vladlen Koltun
Zoran Popovic

Articulated Swimming Creatures
Jie Tan
Yuting Gu
Greg Turk
C. Karen Liu

Locomotion Skills for Simulated Quadrupeds
Stelian Coros
Andrej Karpathy
Ben Jones
Lionel Reveret
Michiel van de Panne

Composite Control of Physically Simulated Characters
Uldarico Muico

Character Animation in Two-player Adversarial Games
Kevin Wampler
Erik Andersen
Evan Herbst
Yongjoon Lee
Zoran Popovic

Color Compatibility from Large Datasets
Peter O’Donovan
Aseem Agarwala
Aaron Hertzmann

Edge-Aware Color Appearance
Min H. Kim
Tobias Ritschel
Jan Kautz

Example-based Image Color and Tone Style Enhancement
Baoyuan Wang
Yizhou Yu
Ying-Qing Xu

Switchable Primaries Using Shiftable Layers of Color Filter Arrays
Behzad Sajadi
Aditi Majumder
Kazuhiro Hiwada
Ramesh Raskar

Contributing Vertices-Based Minkowski Sum of a Non-Convex–Convex Pair of Polyhedra
Hichem Barki
Florence Denis
Florent Dupont

MeshFlow: Interactive Visualization of Mesh Construction Sequences
Jonathan Denning
William Kerr
Fabio Pellacini

LR: Compact Connectivity Representation for Triangle Meshes
Topraj Gurung
Mark Luffel
Peter Lindstrom
Jarek Rossignac

On the Velocity of an Implicit Surface
Jos Stam
Ryan Schmidt

Data-Driven Elastic Models for Cloth: Modeling and Measurement
Huamin Wang
James F. O’Brien
Ravi Ramamoorthi

Frame-based Elastic Models
Benjamin Gilles
Guillaume Bousquet
Francois Faure
Dinesh Pai

Example-Based Elastic Materials
Sebastian Martin
Bernhard Thomaszewski
Eitan Grinspun
Markus Gross

Sparse Meshless Models of Complex Deformable Objects
Francois Faure
Benjamin Gilles
Guillaume Bousquet
Dinesh Pai

Antialiasing Recovery
Lei Yang
Pedro V. Sander
Jason Lawrence
Hugues Hoppe

Local Laplacian Filters: Edge-aware Image Processing with a Laplacian Pyramid
Sylvain Paris
Samuel W. Hasinoff
Jan Kautz

Domain Transform for Edge-Aware Image and Video Processing
Eduardo S. L. Gastal
Manuel M. Oliveira

Non-Rigid Dense Correspondence with Applications for Image Enhancement
Yoav Hacohen
Eli Shechtman
Dan Goldman
Dani Lischinski

Leveraging Motion Capture and 3D Scanning for High-Fidelity Facial Performance Acquisition
Haoda Huang
Jin-Xiang Chai
Xin Tong
Hsiang-Tao Wu

High-Quality Passive Facial Performance Capture Using Anchor Frames
Thabo Beeler
Fabian Hahn
Derek Bradley
Bernd Bickel
Paul Beardsley
Craig Gotsman
Bob Sumner
Markus Gross

Interactive Region-based Linear 3D Face Models
Jose Rafael Tena
Fernando de la Torre
Iain Matthews

Realtime Performance-Based Facial Animation
Thibaut Weise
Sofien Bouaziz
Hao Li
Mark Pauly

Bounded Biharmonic Weights for Real-Time Deformation
Alec Jacobson
Ilya Baran
Jovan Popovic
Olga Sorkine

Blended Intrinsic Maps
Vladimir G. Kim
Yaron Lipman
Thomas Funkhouser

Biharmonic Distance
Yaron Lipman
Raif Rustamov
Thomas Funkhouser

Photo-Inspired Model-Driven 3D Object Modeling
Kai Xu
Hanlin Zheng
Hao Zhang
Daniel Cohen-Or
Ligang Liu
Yueshan Xiong

Two-Scale Particle Simulation
Barbara Solenthaler
Markus Gross

Real-Time Eulerian Water Simulation Using a Restricted Tall Cell Grid
Nuttapong Chentanez
Matthias Mueller

A PML-Based Nonreflective Boundary for Free Surface Fluid Animation
Andreas Soderstrom
Matts Karlsson
Ken Museth

Guide shapes for high resolution naturalistic liquid simulation
Michael Nielsen
Robert Bridson

Animating Fire With Sound
Jeffrey Chadwick
Doug James

Converting 3D Furniture Models to Fabricatable Parts and Connectors
Manfred Lau
Akira Ohgawara
Jun Mitani
Takeo Igarashi

Make it Home: Automatic Optimization of Furniture Arrangement
Lap-Fai Yu
Sai-Kit Yeung
Chi-Keung Tang
Demetri Terzopoulos
Tony F. Chan
Stanley Osher

Interactive Furniture Layout Using Interior Design Guidelines
Paul Merrell
Eric Schkufza
Zeyang Li
Maneesh Agrawala
Vladlen Koltun

Interactive Architectural Modeling with Procedural Extrusions
Tom Kelly
Peter Wonka

Metropolis Procedural Modeling
Jerry O. Talton
Yu Lou
Steve Lesser
Jared Duke
Radomir Mech
Vladlen Koltun

Sensitive Couture for Interactive Garment Design
Nobuyuki Umetani
Danny Kaufman
Takeo Igarashi
Eitan Grinspun

Real-Time Large-Deformation Substructuring
Jernej Barbic
Yili Zhao

Solid Simulation with Oriented Particles
Matthias Mueller
Nuttapong Chentanez

Physics-Inspired Upsampling for Cloth Simulation in Games
Ladislav Kavan
Daniel Gerszewski
Adam Bargteil
Peter-Pike Sloan

Image and Video Upscaling from Local Self-Examples
Gilad Freeman
Raanan Fattal

Scalable and Coherent Video Resizing with Per-Frame Optimization
Yu-Shuen Wang
Jen-Hung Hsiao
Olga Sorkine
Tong-Yee Lee

Subspace Video Stabilization
Feng Liu
Michael Gleicher
Jue Wang
Hailin Jin
Aseem Agarwala

Tonal Stabilization of Video
Zeev Farbman
Dani Lischinski

Making Burr Puzzles from 3D Models
Shi-Qing Xin
Chi-Fu Lai
Chi-Wing Fu
Tien-Tsin Wong
Ying He
Daniel Cohen-Or

A Geometric Study of V-style Pop-ups: Theories and Algorithms
Xian-Ying Li
Tao Ju
Yan Gu
Shi-Min Hu

Depixelizing Pixel Art
Johannes Kopf
Dani Lischinski

Digital Micrography
Ron Maharik
Mikhail Bessmeltsev
Alla Sheffer
Ariel Shamir
Nathan Carr

Computational Stereo Camera System With Programmable Control Loop
Simon Heinzle
Pierre Greisen
David Gallup
Christine Chen
Daniel Saner
Aljoscha Smolic
Andreas Peter Burg
Wojciech Matusik
Markus Gross

Highlighted DOF Photography: Shining Light on Focus
Jaewon Kim
Roarke Horstmeyer
Ig-Jae Kim
Ramesh Raskar

Layered 3D: Tomographic Image Synthesis for Attenuation-based Light Field and High Dynamic Range Displays
Gordon Wetzstein
Douglas Lanman
Wolfgang Heidrich
Ramesh Raskar

A Perceptual Model for Disparity
Piotr Didyk
Tobias Ritschel
Elmar Eisemann
Karol Myszkowski
Hans-Peter Seidel

Circular arc structures
Pengbo Bo
Helmut Pottmann
Martin Kilian
Wenping Wang
Johannes Wallner

Discrete Laplacians on General Polygonal Meshes
Marc Alexa
Max Wardetzky

HOT: Hodge-Optimized Triangulations
Patrick Mullen
Pooran Memari
Fernando de Goes
Mathieu Desbrun

Spin Transformations of Discrete Surfaces
Keenan Crane
Ulrich Pinkall
Peter Schroder

Interactive Editing of Massive Imagery Made Simple: Turning Atlanta into Atlantis
Brian Summa
Giorgio Scorzelli
Ming Jiang
Peer-Timo Bremer
Valerio Pascucci

Geodesic Image and Video Editing
Antonio Criminisi
Toby Sharp
Carsten Rother
Patrick Perez

Matting and Compositing of Transparent and Refractive Objects
Sai-Kit Yeung
Chi-Keung Tang
Michael S. Brown
Sing Bing Kang

Nonlinear Revision Control for Images
Hsiang-Ting Chen
Li-Yi Wei
Chun-Fa Chang

Clipless Dual-Space Bounds for Faster Stochastic Rasterization
Samuli Laine
Timo Aila
Tero Karras
Jaakko Lehtinen

Decoupled Sampling for Graphics Pipelines
Jonathan Ragan-Kelley
Jaakko Lehtinen
Jiawen Chen
Michael Doggett
Fredo Durand

Spark: Modular, Composable Shaders for Graphics Hardware
Tim Foley
Pat Hanrahan

Physically Based Real-Time Lens Flare Rendering
Matthias Hullin
Elmar Eisemann
Hans-Peter Seidel
Sungkil Lee

2011-04-11

How to give a paper presentation

Filed under: Real — liyiwei @ 6:18 pm
Tags: ,

Congratulations that your paper has been accepted to the most prestigious conference in your academic field. You can have a sigh of relief and take a vacation. You deserve it.

But your job is not over yet. You still have to present the paper.

It is important
The presentation is essentially a promo for your paper, and to some degree, for yourself. So it is to your advantage to give a good talk. And if you are relatively junior in the field, you should do it yourself (instead of your adviser or another co-author). This will give you the necessary exposure and recognition.

It is a sales pitch
This can be a personal preference, but for me, a conference talk is more of an advertisement than an education. A very common mistake is trying to cram in all the information into a presentation, hoping that the audience will understand everything. Unfortunately, the talk slots are usually too short and the audience too distracted for this to happen.

Instead, consider the conference as a social event, and your talk a sales pitch. Your goal is to attract people to get interested in your paper, so that they will later spend time reading it. And if you can give a good talk, it also helps sell you as a good researcher. This can translate into long term benefits, like job offers.

Content
The content of your talk should be mainly about the following: (1) why people should care about your paper, (2) why they should read it after your talk, and (3) what are the key points of your paper. Remember, it is more of an advertisement (1 and 2) than education (3).

Actually, there is one more: (4) what a cool and smart guy you are. But you really need to know what you are doing. And if you do, you are beyond this post already.

Practice
Finally, practice your talk, both with yourself and in front of an audience. If you are relatively inexperienced or have a certain degree of glossophobia (most people do), practice enough times so that you can give the talk while sleeping. I practiced at least 20 times for my first SIGGRAPH talk, even though nowadays I usually do just once or twice to avoid cruising through my talk with a too rigid or scripted performance.

2011-01-30

How to review papers

Filed under: Real — liyiwei @ 6:06 pm
Tags: ,

Look for what is good in a paper. For the rest (that is not so good), try to make constructive comments so that the authors can improve the paper.

Looking for the good

I have seen many, many reviews, both as primary/secondary (for which I can see reviewer identities) as well as tertiary (for which I cannot). My overall impression is that on average, reviewers tend to have a negative bias. That is, I have seen more good papers rejected than bad ones accepted. This is sad, as the prosperity and health of the academic community (as well as the human civilization in general) depends crucially on the timely publication of good ideas. The risk/reward ratio is highly asymmetric; a bad idea, even though published, can do little harm as it tend to be quickly ignored and forgotten. (Notable exception: Das Kapital by Karl Marx.) A good idea, if not published, could take tremendous effort to recover (e.g. proving Fermat’s last theorem) or is forever lost (I cannot even give you examples on this).

Why you should be positive as a reviewer

Humans, by nature, tend to be more critical towards others than to themselves. But here are a few reasons why being positive as a reviewer can be good for you:

. This is a small world. What comes around goes around. If you are known as a paper killer, it can not only harm your reputation but also makes others more likely to be negative towards your papers. There are a few well known folks in the graphics community who are highly successful and yet unpopular enough so that they do not tend to be invited to serve as paper committee members. Their opinions also tend to get ignored during committee discussions.

. Your success depends highly on your followers. Do not discourage them. If you see people citing or extending your work, it is to your benefit to help them as they will in turn help you become more famous and more influential in the long run. (Trust me on this: I saw this happened to me before, and I want to see it happen again.) Some people (especially more junior guys) tend to treat this as a kind of threat as they want to claim all credits themselves. The very best researchers start a new field/subject (or at least help popularize it), and move on to a different one leaving others cite and expand their earlier works. People who have to stay on the same subject or field tend to lack sufficient capability or courage to move on.

. The health of a community depends on how the members treat each other. Nobody likes to be in a place where people ambush each other. So do not start to behave like one yourself.

All in all, my experience is that one’s reviews reflect one’s personality and intelligence. Good researchers tend to write good and positive reviews. Cynical and nonconstructive reviews tend to come from those who are sour or feel insecure.

Being fair and constructive

Note that I am not saying you should accept bad papers. Papers with incremental or boring ideas should not be published, as otherwise it is unfair and can drag down the quality of the venue you review for. Papers with good ideas but major flaws should also be fixed before publication, as otherwise they might not attract the attention they deserve. (A common example is a good idea expressed in a bad writing; if people have difficulty understanding an idea, it would probably not make as large an impact it should. In that case, it is usually much better for everyone involved if the writing can be improved before publication.)

Be specific and constructive in your reviews so that the authors could have concrete clues about how to improve their paper.

2011-01-03

Competition

Filed under: Real — liyiwei @ 5:08 pm
Tags: , ,

“Whatever I do, there are always people out there who are smarter, better, tougher, and harder working than me.”

I always keep this in mind and I have found it extremely helpful in keeping me humble and motivated.

I actually never understand why anyone would ever ask me questions like “how hard should I work?” or “how much time should I spend on my research?” Because you will never succeed until you truly appreciate the meaning of the quote above.

2010-12-20

How to use the papers committee list

Filed under: Real — liyiwei @ 11:49 am
Tags: ,

SIGGRAPH now publishes the technical papers committee list a few weeks prior to the submission deadline. Here is my suggestion on how to take this information to your advantage.

First, let me tell you what NOT to do. It is very common for people to try to guess who is likely to review your papers, but this is entirely useless and counterproductive. In general, there is indeed a positive correlation between the committee member expertise and your paper subject, but it is not a very strong correlation and there is a non-trivial chance that a paper will be assigned to an unexpected committee member. (This is mainly due to how the paper sorting and assignment process goes but I am not going to elaborate on this here because I do not think it is necessary for authoring a good paper.)

So my personal strategy is to make sure the paper will survive *any* committee member. This is the safest bet and you cannot go wrong with that. (But do check the committee member list constantly though as sometimes it may change prior to the deadline.) Specifically, you should write your paper (especially the title and abstract) in a way so that it is as clear as possible on what it is all about to minimize the chance of misunderstanding. (If the paper sorters misunderstand your paper, they are more likely to assign your paper to the wrong committee member. And if the senior reviewers misunderstand your paper, they are more likely to assign your paper to the wrong tertiaries. And if reviewers misunderstand your paper they are more likely to score it lowly.)

It also helps to read and cite papers from the committee members for two reasons. The first reason is that the background and expertise of a committee member will obviously have a huge influence on how she is going to review a paper, so if you understand her, you stand a better chance of appreciating how she might approach a paper. The second reason is that psychological studies indicate that it is almost never too much to flatter, and people are usually happy to see their own papers cited and attributed in a positive manner.

If you want to push this further, my personal philosophy is to assume that my paper will be reviewed by the worst possible reviewers and prepare for that. Like Andy Grove said, only the paranoid survive.

« Previous Page — Next Page »

Theme: Rubric. Get a free blog at WordPress.com