Joey Litalien

Phenomenal

I think one subtle thing missing from your diagram—and obviously from the other one too—is the fact that x̂ will not lie exactly on the original data manifold, which is not clear just from the dotted line (maybe the ≈ is enough?). Not sure how to convey this elegantly but love it otherwise!

Fantastic work, congrats!

The dedication 👌

I say "new" but I have not idea when this was rolled out. I just saw it popped today on a few papers.

The new AI outline feature for the Google Scholar PDF reader is pretty sweet. I have not extensively tested it yet but so far it has been quite useful!

I had no idea about the the level-set compositing and meshing system for the original Moana, very cool. Hope more details eventually come out on your newest CSG-based system, it sounds fascinating.

With all the cryptographic tech we have available these days, it’s mindblogging that some administrative processes still require physical documents/signatures. Talk about redtape BS.

I was planning on flying from Tokyo to SF directly after SIGGRAPH to start my new job, but I just learned this week that I absolutely need _physical_ diplomas to apply for TN status at a PoE. Now I have to fly back to Montreal for two pieces of paper. NRT→YUL→YYZ→SFO in 36 hours, wish me luck. 😭

How are Kernel Smoothing in statistics, Data-Adaptive Filters in image processing, and Attention in Machine Learning related?

My goal is not to argue who should get credit for what, but to show a progression of closely related ideas over time and across neighboring fields.

1/n

Wait wait wait, did I miss a RenderMan event? I need that teapot.

SIGGRAPH Asia Tokyo starts tomorrow! I will be presenting our most recent paper on neural warps for product importance sampling. Come say hi at the Sampling & Light Transport session on Thursday @ 1pm! Don’t hesitate to reach out if you want to chat or just hang out. 😄

This is excellent! Really appreciate the effort for the extensive documentation, it will be super useful.

My team RGL is looking for new PhD students starting 2025. If you are excited about topics like inverse rendering, compilers for graphics, and physically based modeling then please join us at EPFL. Info about the lab & admission process: rgl.epfl.ch/pages/jobs (Deadline: Dec 15)

First post on BlueSky and it’s a big one: I have successfully defended my Ph. D. thesis today! Super grateful for my time at the McGill Graphics Lab and the amazing opportunities I’ve had throughout the years. Onto the next adventure! 🙂