Libing Zeng

Working on world model pretraining for robot learning.

[Bernt Bornich @BerntBornich]

Whole body control + NEO looks amazing

[el.cine @EHuanglu]

AI is taking over advertising.

ChatGPT 4o is on FLORA AI canvas now, you can edit image, let anyone wear any outfit, hold any product in video ads and..

use Topaz upscale to 4K

step by step tutorial:

[Jiatao Gu ✈️ COLM2026 @thoma_gu]

In this work, we propose StyleNeRF, a 3D-aware generative model for photo-realistic high-resolution image synthesis with high multi-view consistency, which can be trained on unstructured 2D images.

Joint work with @LingjieLiu1 (MPII) Peng Wang (HKU) and Christian Theobalt (MPII)

[Yuan Liu @YuanLiu41955461]

NeuRay was accepted to CVPR2022!😋
We have updated:
Paper: https://t.co/3sClmTXS2J
Project page: https://t.co/74vX08pf3u
Code: https://t.co/HVeHOXZTWU
Welcome any discussions or questions!

[Jason Y. Zhang @jasonyzhang2]

Tired of taking 100 calibrated images to get a reasonable 3D reconstruction? Our new #NeurIPS2021 paper recovers shape, texture, and illumination of 3D objects from ~8 images in the wild without precise camera poses! https://t.co/IM7qjUZgrc

[AK @_akhaliq]

TensoRF: Tensorial Radiance Fields
abs: https://t.co/keoZNR20v0
project page: https://t.co/Dy6PxjxZ5c

[Jon Barron @jon_barron]

Very glad I can finally talk about our newly-minted #CVPR2022 paper. We extended mip-NeRF to handle unbounded "360" scenes, and it got us ~photorealistic renderings and beautiful depth maps. Explainer video: https://t.co/miHACOCOcG and paper: https://t.co/5K99BZDE8d

[Jon Barron @jon_barron]

Training NeRFs per-scene is so 2020. Inspired by image based rendering, IBRNet does amortized inference for view synthesis by learning how to look at input images at render time. 15% drop in error, 80% fewer FLOPs than NeRF. Great work @QianqianWang5! https://t.co/y0hCJmeHlJ

[AK @_akhaliq]

State-of-the-Art in the Architecture, Methods and Applications of StyleGAN
abs: https://t.co/SRnVrtfU7T

[Rinon Gal @RinonGal]

The Nicolas Cage version of #StyleGAN3-NADA is coming along quite nicely🙃

[Li-Yi Wei @liyiwei]

I thought Facebook would be renamed to "Faceverse" https://t.co/xZkPY3dJ9o 😁

[Justin Johnson @jcjohnss]

Single-image view synthesis usually handles just tiny viewpoint changes. Our new paper goes far beyond the input view by combining differentiable point cloud rendering (for 3D awareness) and vector-quantized autoregressive modeling (for extrapolation)

[Seoyeon Stella Yang @codestella]

CLIP+NeRF: DietNeRF🥗 is now available! *With only 8 images*, DietNeRF is capable of neural rendering quality novel view from a 3D model. You can check out more details here:
👉𝐆𝐢𝐭𝐡𝐮𝐛 : https://t.co/TONkoiQuSd
👉𝐒𝐭𝐞𝐚𝐦𝐥𝐢𝐭 𝐒𝐩𝐚𝐜𝐞 𝐃𝐞𝐦𝐨 :
https://t.co/H6sdvJtHqY

[AK @_akhaliq]

ADOP: Approximate Differentiable One-Pixel Point Rendering
abs: https://t.co/npOqsAstAx

[AK @_akhaliq]

stylegan3 is out
github: https://t.co/ctkX7uzSNM

[Prune Truong @prunetruong]

📯 ICCV 2021 Oral: "Warp Consistency for Unsupervised Learning of Dense Correspondences", with @MDanelljan, @DrFisherYu and Luc Van Gool.

Join us Session 8, 13th 9AM and 15th 4PM (EDT) if you want to know more!

paper: https://t.co/FAOPl9MIg3
video: https://t.co/N3AZq9XHoC

[Jia-Bin Huang @jbhuang0604]

Deep 3D Mask Volume for View Synthesis of Dynamic Scenes

• Create a multi-view video dataset
• Stereo video ➡️ 3D mask volume (disoccluded regions) ➡️ blend static/dynamic MPI.

https://t.co/GYaocjYV41

[Elliott / Shangzhe Wu @elliottszwu]

Announcing @ICCV_2021 workshop on "Unsupervised 3D Learning in the Wild" with an incredible line-up of speakers on this topic! #ICCV2021

🚩 Website: https://t.co/uqQv7iKyKs
📅 Time: 7:00-18:00 EDT / 12:00-23:00 BST, 11 Oct 2021
Calendar: https://t.co/W0a5IrOJRV (mark it down!)

[darthgera123 @PulkitRega]

Mind-blowing work

[AK @_akhaliq]

TöRF: Time-of-Flight Radiance Fields for Dynamic Scene View Synthesis
pdf: https://t.co/90qPiXNs6J
abs: https://t.co/ACkdet9ynV
project page: https://t.co/XAOju9kPYS