HN · GitHub Signals

OPEN SOURCE, IN FOCUS

seen2scene

quan-meng/seen2scene

[ ECCV 2026 ] Seen2Scene takes an incomplete real-world 3D scan and generates a complete, coherent 3D scene using visibility-guided flow matching — trained directly on real-world data.

Explore on GitHub ↗

Saved projects stay in this browser. No account needed.

TAKE A CLOSER LOOK

Start with the source.

The repository is the place to check installation steps, supported platforms, the license, and current maintenance before choosing a tool.

Read the documentation on GitHub ↗

Keep exploring

Find projects in the same areas, or search the archive for a different approach.

ANOTHER GOOD FIND

Explore related projects

View the archive ↗

Connected by topics and descriptions.

nicolasvonluetzow

gaussiangpt

[ECCV'26] Our method creates 3D Gaussian scenes completely autoregressively, allowing for generation, completion, and outpainting with the same model.

Take a look →
jaceyhuang

gen3r

[CVPR 2026] Gen3R: 3D Scene Generation Meets Feed-Forward Reconstruction

Take a look →
powertony102

recal3r

ReCal3R: Reliability-Calibrated Learning Rates for Streaming 3D Reconstruction

Take a look →