Mike Wong
@artixels
Followers
2K
Following
9K
Media
398
Statuses
1K
pixel surgeon
latentpark
Joined June 2009
Splannequin: Freezing Monocular Mannequin-Challenge Footage with Dual-Detection Splatting TL;DR: How to freeze a dynamic molecular scene when people inevitably move. Contributions: • A Novel Problem Formulation and Benchmark: We are the first to formally address synthesizing
1
4
39
今年は2次元のtransient renderingを使った作品を提出しました。 運営の皆様お疲れ様でした。 #レイトレ合宿
3
106
702
The same stippling method can create some interesting depth of field effect.
Found a Depth of Field demo code of 2.5D moving stipples; here we slide the image plane through Z. #bluenoise #LagrangianImage #atoMeow
0
1
10
Wow, a *phenomenal* single author paper by @ryichando
https://t.co/qS7GZqKWnT
Every cape is one bug away from becoming a scarf. 🦸♂️ But one scientist finally fixed it - millions of ribbons, noodles, and fabrics twisting together. Human ingenuity at its best! What a time to be alive! Full video: https://t.co/kqrnLYt7Hf
1
0
3
My talk from @BetterSoftwareC last week is up on youtube. I present my findings on thread synchronization and job systems that I learned while parallelizing the physics solver.
16
95
731
🔥😍🔥
I'm thrilled to share the VQVDB trailer 😎 More info : https://t.co/NdAdMKApJM
#houdini #ai #ml #openvdb
0
0
0
artistic exploration via sampling a tiny neural network's memory. curate your own data and train your own model. #diffusion #neural #generative
1
0
5
Selected more photos from my collection and finished a better dataset for my personal cloudscape model training. At the end, it's the quality of data that matters most for any machine learning solution.
0
0
1
Spent some time going outdoor to shoot more cloud photos. I quickly used around 100 crops and trained a tiny model. Quite happy with the results.
0
0
2
The season finale of Lighter Darker is loads of fun and I am a proud owner of a *Physical* copy of Knoll Light Factory but I really want to try Knoll’s Flare Box
The first season of the ILM podcast comes to a close with our Lens Flare Spectactular. We're joined by John Knoll and Shannon Tindle to talk about our favorite lens flares from movies, the favorite flares we've made, and much much more. Listen here: https://t.co/ZvoTB5tFEU
0
0
3
Right-sizing a diffusion model seems important for capturing the different levels of variation of a target distribution. This particular generated sample truly captures what I like about the training set. There are so many dimensions of art making with Machine Learning.
0
0
0
Attention is *not always* our need Removed the Attention modules from the UNet Halved the filter kernel numbers A more compact UNet learned faster and an improved distribution, i.e. better unseen samples.
0
0
3
A quick test of using the tiny diffusion model trained last week for 2K generation. The result is meaningful and reasonable, likely because of the data distribution nature of cloud images. It's possible to create a realtime infinite scrolling panorama with this tiny model.
1
1
3
Trying to see how a tiny diffusion model will mix or separate the distribution of two flower images during training. This one is a nice in-between.
0
0
0
Training a tiny diffusion model using my previous work (12 images) as a dataset. The ability to sample and visualize a distribution of your own aesthetics is pretty cool. I am using the elegant Iterative 𝛼-(de)Blending method (IADB) proposed by Heitz et al (see reply)
1
0
1
only a few epochs and the samples feel a bit impressionistic, kind of nice.
0
0
0