I’m particularly appreciative of the support I’ve gotten from @CorinnaCortes, Katherine Chou, @ahfdc, @StephenWrage, Reuben Brigety, and more recently, my pre-doctoral scientific advisors, @kiamada and @WonderMicky.
Rethinking image representations for autoregressive generation! Our new Spectral Image Tokenizer works in the image spectrum, where the coarse-to-fine representation has a more natural sequential interpretation
Our new paper, "Spectral Image Tokenizer", is on arXiv! We train a tokenizer on DWT coefficients
that enables autoregressive coarse-to-fine image generation, w/ applications to multiscale text-to-image, and text-guided editing.
w/ @kiamada, @msuhail153 https://t.co/0s4KYduBRx
@TacoCohen ‘s first equivariant paper got me into this wonderful journey of symmetry. Here, at the very place where he received the Best Paper award for Spherical Networks in 2018!
This was an amazing presentation by @erikjbekkers on equivariant neural fields and neural ideograms at the Equivision Workshop. I want to go and read all his 2024 papers now!
At #CVPR2024, I will give a talk about "Geometric Deep Learning for Weather" at the Equivariant Vision workshop Tue 2pm https://t.co/lz4fzCZOtK, and I'll present a poster on Single Mesh Diffusion Wed 5pm https://t.co/lQV19vg41z w/ @twmitchel and @kiamada. Hope to see you there!
Giving a short talk today at the C3DV workshop today at CVPR around 2:20 PM. Will be discussing our main conference paper “Single Mesh Diffusion with Field Latents for Texture Generation”. 1/2
Our Equivariant Vision workshop features five great speakers @erikjbekkers@HaggaiMaron@ninamiolane@_machc, and Leo Guibas, spotlight talks, posters, and a tutorial prepared for the vision audience. Come tomorrow, Tuesday, at 8:30am in Summit 321! Thank you @CongyueD for leading the organization!
https://t.co/tp7HaCoa4H
Generate high quality textures with single mesh LDMs!
#CVPR2024
Our *intrinsic* 3D diffusion models, trained on a single mesh, can generate texture variations, perform inpainting, and even transfer textures to different shapes.
https://t.co/SQ9WcF5DdE
w/@twmitchel & @_machc
An image reconstructive representation learning loss that keeps scaling !!! You know how MAE does not benefit well from XLarge data. Well, if you borrow lessons from autoregressive language modeling, you can learn from Billion images and get 84% on IN1K. W/ Apple ML colleagues.
Would you like to know how to define and realize an equivariant convolution and attention on a light field / feature field? Come to listen to @xu_yinshua86846 and @JiahuiLei1998 presenting our spotlight at 10:45am at poster#309 @NeurIPSConf
The first major WeatherBench 2 update is now live, including:
➕ New models: (pseudo-)operational Pangu and GraphCast, FuXi, SphericalCNN and NeuralGCM
➕ Improved scorecards (incl. a probabilistic one)
➕ More comprehensive ERA5 and IFS ENS data
https://t.co/f50rGD3GcK
🌟FisherRF introduces a novel approach for view selection & uncertainty in 3D Gaussian Splatting & NeRFs. Fisher information can be computed without altering model architecture, and it is as cheap as back-propagation! 🚀More details: https://t.co/ULWTDTI7y2
#CVPR2024 Such a great effort by my wonderful students. Excellent finish. Good luck to all of you. What a great feeling, 31 years of CVPR deadlines :-)
CVPR 1999 Best Paper Award
Robust Hierarchical Algorithm for Constructing a Mosaic from Images of the Curved Human Retina
A. Can, C. V. Stewart, B. Roysam
#TBThursday
We proudly present our 524 page book on equivariant convolutional networks.
Coauthored by Patrick Forré, @erikverlinde and @wellingmax.
https://t.co/y9YBpqhyLG
[1/N]