• Interactive Poisson Blending on GPU
    Interactive Poisson Blending on GPU
  • 404 Not Found In Two Triangles
    404 Not Found In Two Triangles
  • Simplest and Fastest GLSL Edge Detection using Fwidth
    Simplest and Fastest GLSL Edge Detection using Fwidth

Social Street View wins the Best Paper Award in Web3D 2016

I am thrilled to announce that Social Street View has won the Best Paper Award at Web3D 2016, which was held next to SIGGRAPH 2016 in late August! Please visit the project website for more details: http://www.socialstreetview.com I also published my slides and videos.  


Re: Levels of reality

I tried to retain the habit of posting something per day, as inspired by Ken Perlin. If I cannot find a good topic, I turn to his blog. E.g. in http://blog.kenperlin.com/?p=23712, Ken discussed “Could we create a hierarchy of fictional realities, through some reasonably objective criteria? If so, what would those criteria be?” It is…


References Management

I used Mendeley 7 years ago while working at HCIL. One of the most annoying things is to capitalize titles, correct DOIs, and rename filenames. Thanks to Python and many, many package contributors, I could finally automatic this process to some extent using https://github.com/ruofeidu/DuWebKit/blob/master/scripts/utils/autobib.py and https://github.com/ruofeidu/DuBibtex 🙂


Free Guy, Geollery, and Metaverse

Metaverse might be the #1 IT buff word emerged during the global pandemic, but what if we could actually create one? The movie “Free Guy” depicts a fantasy world, where an NPC could evolve with real intelligence, save the code and world which was stolen by evil CEOs from poor, little creators / coders /…


[Summary] Google I/O and Microsoft Build 2018

Today, two technology festivals were held in Mountain View and Redmond respectively: Google I/O and Microsoft Build 2018. In summary, both companies have all in AI. Google published email auto-completion, photo auto-spot, auto-colorization, better sound synthesis, memorable Q&A, Android P, text from images, style match. Microsoft presented its Fluent Design, Azure AI-enabled edge devices (phone,…


[Summary] StackGAN: Text to Photo-realistic Image Synthesis

StackGAN Results

StackGAN has got nearly 200 citations since first appeared on arXiv on December 10, 2016.  The StackGAN is the first to generate 256*256 image with photo-realistic details from text description.   Generative Adversarial Network (GAN), originally proposed by Ian. It takes advantage of a generator network and a discriminator. The generator is trained to fool…


[Summary] Omnipresence 3D for Multiview Mixed Reality

Christian Laforte leads the development of Fortem’s Omnipresence 3D software. It helps organizations prevent costly mistakes and respond to incidents in record time — protecting our communities against increasingly complex security threats. This is definitely a great leap of my prior work, VideoFields. At the time, I was offered only three surveillance video cameras and would…


[Summary] PointNet, PointNet++, and PU-Net

Shuo Li presented a talk at GVIL weekly seminar: PointNet, PointNet++, and PU-Net PointNet Instead of 3D convolution, PointNet directly consumes point clouds, which well respects the permutation invariance of points in the input. A point cloud is an unordered set of vectors. Each point Pi is a vector of its (x, y, z) coordinate plus extra…


Gradient, Circulation, Laplacian, Divergence, Jacobian, Hessian, and Trace

Here is a summary of all these concepts. Most the concepts are from Wikipedia. A larger figure is shown below: In mathematics, the gradient is a multi-variable generalization of the derivative. While a derivative can be defined on functions of a single variable, for functions of several variables, the gradient takes its place. The gradient…


[Summary] Talk by Dr. Chakareski: Networked Virtual and Augmented Reality: The New Frontier

Jacob Chakareski is an Assistant Professor of Electrical and Computer Engineering at The University of Alabama, where he leads the Laboratory for VR/AR Immersive Communication (LION). His interests span networked virtual and augmented reality systems, UAV-IoT sensing and communication, and rigorous machine learning for stochastic control. Dr. Chakareski received the Adobe Digital Experience Faculty Research…


Estimated Cost of Per Atom Function in Real-time Shaders on the GPU

I have created a grouped list of estimated cost of instructions according to this Chinese reference. This may not be accurate, but is mostly correct from my experience. Some intuitions are: Abs, saturate are free (Why is clamp in GLSL not free? I doubt it) Log, exp, sqrt are almost free! (That’s why Kernel Foveated Rendering…