Lightnews — Scholar-powered news

Baptiste GENEST @baptiste-genest.bsky.social · 7d

I could try between points embeded in high dimensions with the spectral embedding we use to sample meshes for instance

Baptiste GENEST @baptiste-genest.bsky.social · 7d

We had experiments using the toy datasets of some ML papers (between gaussians for example) in dimension up to 20 (and it behave well!) but we mainly lacked interesting data, if you have ideas or references to get some in that kind of dimensions that would be great 🙏

2

Baptiste GENEST @baptiste-genest.bsky.social · 7d

Of course! That would be awesome!

The code interface is very simple (at least in the bijective case).

We mainly tested BSP-OT on graphics applications in low dimension, I hope that POT users would still find it usefull !

1

Reposted by Baptiste GENEST

Nicolas Bonneel @nbonneel.bsky.social · 7d

Happy to share the excellent work of @baptiste-genest.bsky.social on making sliced optimal transport *much* better by replacing the sort along a line with a variant of QuickSort that directly takes multiples directions at once! It also directly provides a coupling between points.

Baptiste GENEST @baptiste-genest.bsky.social · 7d

Computing the exact bijection of the optimal transport (OT) problem between very large point sets is completely untractable…

In our SIGGRAPH Asia 2025 paper: “BSP-OT: Sparse transport plans between discrete measures in log-linear time” we get one with typically 1% of error in a few seconds on CPU!

1 7

Reposted by Baptiste GENEST

David Coeurjolly @dcoeurjo.bsky.social · 7d

Glad to share with you our last paper to be presented @ SIGGRAPH Asia 2025 !
Kudos @baptiste-genest.bsky.social

Baptiste GENEST @baptiste-genest.bsky.social · 7d

Computing the exact bijection of the optimal transport (OT) problem between very large point sets is completely untractable…

In our SIGGRAPH Asia 2025 paper: “BSP-OT: Sparse transport plans between discrete measures in log-linear time” we get one with typically 1% of error in a few seconds on CPU!

1 7

Baptiste GENEST @baptiste-genest.bsky.social · 7d

Check the paper : baptiste-genest.github.io/papers/bsp-o...

Code : github.com/baptiste-gen...

I will present this work during the conference! (15-18 december)

GitHub - baptiste-genest/BSP-OT: Official source code of the "BSP-OT: Sparse transport plans between discrete measures in loglinear time" paper (SIGGRAPH Asia 2025).

Official source code of the "BSP-OT: Sparse transport plans between discrete measures in loglinear time" paper (SIGGRAPH Asia 2025). - baptiste-genest/BSP-OT

github.com

1 3

Baptiste GENEST @baptiste-genest.bsky.social · 7d

We show that this approach can be applied to standard applications of OT, like image stippling, shape interpolation, color transfer, partial point cloud registration, intrinsic mesh sampling and Topological Data Analysis (a domain where bijectivity is required!).

bijective shape interpolation, comparaison with geomloss

1 4

Baptiste GENEST @baptiste-genest.bsky.social · 7d

We also extend it to partial matchings, and general sparse transport plans between non-uniform measures.

Then, in the bijective and injective cases, we exploit this extreme sparsity to efficiently merge two assignments using very simple discrete optimization.

local assignment swapping, only consider 4 points

global merging : select best assignment on each connected component of the union

1 3

Baptiste GENEST @baptiste-genest.bsky.social · 7d

Remarkably, this procedure is also a generalization of the (projection + 1D assignment) step used at the core of Sliced OT! The difference being that it preserves much more information than a single projection, which yields couplings of much higher quality, with the same complexity.

1 5

Baptiste GENEST @baptiste-genest.bsky.social · 7d

BSP Matchings generalize the Quicksort algorithm. We simultaneously build two BSP trees on each point cloud and match their leaves.

BSP matching : recursive splitting of the points along linear slices, hence matching the leaves of two BSPs

1 1 4

Baptiste GENEST @baptiste-genest.bsky.social · 7d

This is a joint work with @nbonneel.bsky.social, Vincent Nivoliers and @dcoeurjo.bsky.social @cnrsinformatics.bsky.social

Our pipeline is simple: we use BSP (Binary Space Partitioning) matchings to efficiently generate random bijections before merging them into a single one of low transport cost.

1 4

Baptiste GENEST @baptiste-genest.bsky.social · 7d

Computing the exact bijection of the optimal transport (OT) problem between very large point sets is completely untractable…

In our SIGGRAPH Asia 2025 paper: “BSP-OT: Sparse transport plans between discrete measures in log-linear time” we get one with typically 1% of error in a few seconds on CPU!

1 12 34

Reposted by Baptiste GENEST

Clément Canonne @ccanonne.github.io · Aug 19

"The greatest trick the Binomial ever pulled was convincing the world it didn't behave exactly as expected."

If N is a Binomial random variable with parameters n and p, then the expected value of "N choose k" is "n choose k" time p raised to the k-th power.

2 1 18

Reposted by Baptiste GENEST

laydgeur @laydgeur.bsky.social · Aug 15

Vous avez sûrement déjà lu que la climatisation aggrave les canicules car l’air chaud rejeté à l’extérieur augmenterait la température des villes de 2 à 3°C.

Vérifions ce que disent vraiment les études scientifiques à ce sujet 🧑‍🔬

(lisez jusqu’au bout, car vous allez être 🤯🤯 à la fin)

#Thread
⬇️⬇️

Carte de Paris simulation îlot de chaleur

82 710 1.3K

Reposted by Baptiste GENEST

Axel Paris @axelparis.bsky.social · Aug 5

📜 New SIGGRAPH 2025 paper 🎉

❔How to compute bounding volumes for procedural Signed Distance Fields (SDFs)? This is not so trivial!

💡We propose a simple method called Sphere Carving. It extracts (convex) bounding volumes around SDFs, requires very few evaluations, and is GPU compatible.

1 11 27

Reposted by Baptiste GENEST

Craig Reynolds @craigreynolds.bsky.social · Jul 29

OPEN SCIENCE IS THE FUTURE!

Very pleased to see the Association for Computing Machinery (ACM — the parent organization to my “home” conference SIGGRAPH) has announced it is transitioning to full open access by January 2026!!

#ACM #open #access #science #publication #journal #compsci

Open Access Publication & ACM

www.acm.org

1 20 70

Reposted by Baptiste GENEST

Wilhem Barbier @wbrbr.bsky.social · Jun 23

I am glad to attend #HPG2025 where I just presented our paper "Fused Collapsing for Wide BVH Construction", co-authored with Mathias Paulin. We propose a fast build algorithm for wide BVHs that directly computes a wide hierarchy without additional collapsing pass. Webpage: wbrbr.org/publications...

Fused Collapsing for Wide BVH Construction

wbrbr.org

1 12 44

Baptiste GENEST @baptiste-genest.bsky.social · May 16

Thank you so much for your feedback 🤗🤗🤗 I was very lucky, the ideas were very simple and worked very quickly. From the first ideas to the submission it took around 4 months and a half🙏

2

Baptiste GENEST @baptiste-genest.bsky.social · May 13

Thank you so much for your comment🙏🙏 it means the world to us🤗

1

Baptiste GENEST @baptiste-genest.bsky.social · May 11

Thank you! It's really nice of you🙏

1

Baptiste GENEST @baptiste-genest.bsky.social · May 10

It was made during an intership at @adobe.com
. This is a joint work with Pierre Gueth, Jérémy Levallois and Stephanie Wang, who supervised me during my internship. Many thanks to them!

Here is the submitted version of the article: baptiste-genest.github.io/papers/impli...

baptiste-genest.github.io

4

Baptiste GENEST @baptiste-genest.bsky.social · May 10

You can then use these coordinates to enhance SDF rendering with textures, normal maps, and even displacement maps!

You also can use it with shell maps to define a 3D coordinate system from the 2D ones, and more!

1

Baptiste GENEST @baptiste-genest.bsky.social · May 10

We use concepts from differential geometry to first define local uv-fields that are evaluatable point-wise. We then show how we can merge multiple fields into larger ones to get high quality uv-fields on large smooth patches of the shape.

1 1

Baptiste GENEST @baptiste-genest.bsky.social · May 10

We introduce the first method to compute uv coordinates on implicit surfaces, without discretization, in real-time in a sphere-tracing pipeline!

It allows you to get full control over the appearance of the shape, with textures, without losing the benefits of implicit representations!

1 1

Baptiste GENEST @baptiste-genest.bsky.social · May 10

Implicit surfaces are great, but if you are not Inigo Quilez, it's really hard to control how they look...

That's why I'm really proud to annonce our Eurographics 2025 paper: "Implicit UVs: Real-time semi-global parameterization of implicit surfaces".

4 15 54