Advertisement ยท 728 ร— 90

Posts by Julian Ost

Post image Post image Post image Post image

๐—ช๐—ผ๐—ฟ๐—น๐—ฑ๐—™๐—น๐—ผ๐˜„๐Ÿฏ๐——: ๐—™๐—น๐—ผ๐˜„๐—ถ๐—ป๐—ด ๐—ง๐—ต๐—ฟ๐—ผ๐˜‚๐—ด๐—ต ๐Ÿฏ๐—— ๐——๐—ถ๐˜€๐˜๐—ฟ๐—ถ๐—ฏ๐˜‚๐˜๐—ถ๐—ผ๐—ป๐˜€ ๐—ณ๐—ผ๐—ฟ ๐—จ๐—ป๐—ฏ๐—ผ๐˜‚๐—ป๐—ฑ๐—ฒ๐—ฑ ๐—ช๐—ผ๐—ฟ๐—น๐—ฑ ๐—š๐—ฒ๐—ป๐—ฒ๐—ฟ๐—ฎ๐˜๐—ถ๐—ผ๐—ป
Amogh Joshi, Julian Ost, Felix Heide
arxiv.org/abs/2603.29089
Trending on www.scholar-inbox.com

2 weeks ago 3 1 0 0
Video

WorldFlow3D ๐ŸŒ by Flow Through Hierarchical Distributions, without VAEs! We reformulate 3D generation as flowing through sequentially finer 3D distributions, cutting training time by more than half! Vectorized maps provide full controllability Project: worldflow3d.com, with @amogh7joshi.bsky.social

2 weeks ago 0 0 0 0
Video

Looking for fully configurable 3D street scene assets and real-time rendered videos? Our latest work generates physically-grounded 3D scenes ideal for robot learning & testing.

Check out our paper + interactive demo: light.princeton.edu/lsd-3d

7 months ago 2 1 0 0
Video

Super excited to share our @natmachintell.nature.com paper (www.nature.com/articles/s42...), in which we recast and generalize 3D tracking as an inverse neural rendering task where we fit a scene graph to an image that best explains this image.

Project and Paper: light.princeton.edu/publication/...

7 months ago 20 3 0 0
Preview
AI method reconstructs 3D scene details from simulated images using inverse rendering Over the past decades, computer scientists have developed many computational tools that can analyze and interpret images.

A new AI-based inverse rendering method reconstructs 3D scene details from simulated images, offering improved generalization and transparency across diverse real-world datasets. doi.org/g9x2cv

7 months ago 2 1 0 0
Preview
Neural Fields Beyond Conventional Cameras Neural Fields Beyond Conventional Cameras

Only a couple weeks left to submit a paper to Neural Fields Beyond Conventional Cameras at CVPR 2025! neural-bcc.github.io#call4paper

Our *non-archival* workshop welcomes both previously published and novel work. A great opportunity to get project feedback and connect with other researchers!

1 year ago 4 2 0 1

Congrats ๐Ÿฅณ

1 year ago 5 0 1 0
Post image

Following an excellent debut at ECCV 2024, we're excited to announce the 2nd Workshop on Neural Fields Beyond Conventional Cameras at this CVPR 2025 in Nashville, Tennessee!

Workshop site: neural-bcc.github.io

1 year ago 8 7 1 1
Advertisement
Preview
a car is driving through a parking lot with a store in the background ALT: a car is driving through a parking lot with a store in the background

If youโ€™re going to flag a missing reference to an arXiv paper published after the #CVPR2025 submission deadline as a major weakness, could you also provide access to one of these? ๐Ÿ‘‡

1 year ago 10 1 1 0

For other 3D vision newcomers to blue sky: I highly recommend joning @chrisoffner3d.bsky.social 's list to follow the right people ;): go.bsky.app/Cfm9XFe

1 year ago 19 4 2 0