Suyeon Choi retweeted
Optics as code - a new paradigm in physical AI
Optical neural networks are notoriously hard to calibrate and suffer from the sim2real gap. We show that a coding agent with the hardware in the loop quickly tunes the optics and improves image classification from 35% to 96.8%.
🧵1/6
Suyeon Choi retweeted
Together with our collaborators at Stanford, I’m proud to share the publication of our latest Nature Photonics article. This paper builds on our past work on Holocake and Waveguide Holography to advance our goal to create VR glasses.
computationalimaging.org/pub…
Suyeon Choi retweeted
🚀 Just published in Nature Photonics: synthetic aperture waveguide holography—a new path toward ultra-thin, high-quality 3D mixed reality displays.
📄 nature.com/articles/s41566-0…
#Photonics #Holography #MR
1/5
Suyeon Choi retweeted
Excited to share our new #SIGGRAPH2025 paper! In this work, we show how to combine Gaussian splatting and computer-generated holography using Gaussian Wave Splatting. This enables photorealistic 3D holograms for emerging holographic VR/AR displays.
1/4
Suyeon Choi retweeted
Hi there, we at #HKU are looking for Postdocs/PhD students in the interdisciplinary research field of optics and computer science. Check here for details eee.hku.hk/~evanpeng/images/… Thx for helping spread out, cheers~
Suyeon Choi retweeted
Excited to share our new #SIGGRAPHAsia2024 paper on étendue expansion & light field holograms generation for holographic displays using a multi-source laser array! (1/n)
Suyeon Choi retweeted
Excited to share our new Nature paper! In this work, we propose a new display design that pairs inverse-designed metasurface waveguides with AI-driven holographic displays to enable full-color 3D augmented reality from a compact eyeglasses-like form factor.
1/8
Suyeon Choi retweeted
Apple announced Vision Pro yesterday. Just like the iPhone, this system is poised to disrupt how we interact with digital information. Interestingly, perhaps not surprisingly, eye tracking is key to enabling future human-computer interfaces in their mixed reality system. (1/n)
Suyeon Choi retweeted
We are excited to present our work at #NeurIPS22 on Generative Neural Articulated Radiance Fields! In this paper, we build on top of the capabilities of existing 3D GAN architectures and propose the first method for generating full human bodies! (1/7)
Suyeon Choi retweeted
Announcing the Toronto Computational Imaging Group @UofT! I'm excited to join Kyros Kutulakos and Roman Genov in leading the group. Our research is at the intersection of imaging, sensors, computer vision, and machine learning. And we're hiring students!
compimaging.cs.toronto.edu
Suyeon Choi retweeted
*Mitsuba 3* is now available! It's a major redesign of the lab's infrastructure for differentiable rendering building on the Dr.Jit just-in-time compiler announced yesterday. Full video link: tinyurl.com/mitsuba-tut 1/10
Suyeon Choi retweeted
Physics vs. Learned Priors: Rethinking Camera and Algorithm Design for Task-Specific Imaging.
arxiv.org/pdf/2204.09871.pdf
I find it an instructive survey paper from @raskarmit's presenting a taxonomy of modern computational imaging tech. that include learning, physics models & HW.
Suyeon Choi retweeted
I’m so excited to share this glimpse into our research towards passing the visual Turing test. @nchan covers the latest on varifocal, retinal resolution, distortion, HDR, Holocake, and more. Several prototypes will be shown at SIGGRAPH 2022, including a demo of our HDR headset.
we went up to Reality Labs Research to get exclusive hands-on time with Meta’s newly announced VR headset prototypes! youtu.be/x6AOwDttBsc
Suyeon Choi retweeted
We released the code and models for our upcoming #CVPR2022 paper EG3D for unsupervised generation of 3D-aware images and geometry from single-view 2D image collections github.com/NVlabs/eg3d
Efficient Geometry-aware 3D Generative Adversarial Networks
abs: arxiv.org/abs/2112.07945
project page: matthew-a-chan.github.io/EG3…
demonstrate sota 3D-aware synthesis with FFHQ and AFHQ Cats, among other experiments
Suyeon Choi retweeted
Today we’d like to highlight features from functorch, a beta PyTorch library that provides JAX-inspired function transformations like vmap. (pytorch.org/functorch/)
If you’re not sure what sort of cool new things vmap allows you to do, read on to learn more!
(1/n)
Suyeon Choi retweeted
NVIDIA Researchers Demonstrate Ultra-thin Holographic VR Glasses That Could Reach 120° Field-of-view roadtovr.com/nvidia-stanford… @jkim_kr_
Suyeon Choi retweeted
Ecstatic to share our work on using next-generation focal-plane sensor—processors for motion deblurring! Thanks @ak92501 for tweeting about us! 1/n
Learning Spatially Varying Pixel Exposures for Motion Deblurring
abs: arxiv.org/abs/2204.07267
project page: ccnguyen.github.io/lsvpe/
Suyeon Choi retweeted
Publication: Towards Practical Physical-Optics Rendering
Physical light transport made practical, with importance-sampled wave BSDFs, ray traced physical-optics beams and derived limits on the transformation of optical coherence.
SIGGRAPH22 (to appear) ssteinberg.xyz/2022/04/03/pr…
Suyeon Choi retweeted
Pleased to announce our #CVPR2022 paper on BACON: band-limited coordinate networks! This is an interpretable network architecture with a controllable Fourier spectrum for multiscale scene representation.
video: youtube.com/watch?v=zIH3KUCg…
paper: arxiv.org/abs/2112.04645
1/n
Suyeon Choi retweeted
Block-NeRF: Scalable Large Scene Neural View Synthesis
abs: arxiv.org/abs/2202.05263
project page: waymo.com/research/block-ner…





















