Applied math and Geometry lover, discrete geometer wannabe. PhD Student at LIRIS. he/him baptiste-genest.github.iohttps://bsky.app/profile/baptiste-genest.bsky.socialhttps://bsky.app/profile/baptiste-genest.bsky.social/post/3msq4aruxkk2lI know nothing about GPU programming, so I thought that BSP-OT on the GPU would not work because we are recursive then sequential... I asked Claude to give it a try in CUDA: x10 speed-up 😳 0.4s for 350k points in 3D, with the same quality as the CPU 😱 Here is the code: https://github.com/baptiste-genest/BSP-OT-CUDA/tree/main [contains quote post or other embedded content]10 Aug 2026 12:32 +0000at://did:plc:ia2a55zrh22gor73moijadrp/app.bsky.feed.post/3msq4aruxkk2lhttps://bsky.app/profile/baptiste-genest.bsky.social/post/3mrkiu7dnoc25At the end of my internship at Adobe, we could not release code for numerous reasons. But I thought that the pseudo-code in the paper would be enough for anyone to re-implement it! I was wrong 🫠 @arezdev.de contacted me a week ago, asking why his implementation did not work... [contains quote post or other embedded content]26 Jul 2026 13:37 +0000at://did:plc:ia2a55zrh22gor73moijadrp/app.bsky.feed.post/3mrkiu7dnoc25https://bsky.app/profile/baptiste-genest.bsky.social/post/3mnpc6j3ms22aI'm *thrilled* to announce that: "MORE pictures of my cat" have been accepted for publication on *my website* 😻! baptiste-genest.github.io/toffee/ This is joint work with my girlfriend👏07 Jun 2026 13:40 +0000at://did:plc:ia2a55zrh22gor73moijadrp/app.bsky.feed.post/3mnpc6j3ms22ahttps://bsky.app/profile/baptiste-genest.bsky.social/post/3mnp6xgouak2mSuper happy and proud of this! Many thanks to @rflamary.bsky.social for his time and help 🙏 I'd strongly encourage anyone to contribute to the library, Rémy makes it easy! Check out the example: https://pythonot.github.io/master/auto_examples/plot_bsp_ot.html ! [contains quote post or other embedded content]07 Jun 2026 12:42 +0000at://did:plc:ia2a55zrh22gor73moijadrp/app.bsky.feed.post/3mnp6xgouak2mhttps://bsky.app/profile/baptiste-genest.bsky.social/post/3mmtm4tzb7225I'm extremely grateful to the award committee and to my awesome advisor @dcoeurjo.bsky.social ! [contains quote post or other embedded content]27 May 2026 13:23 +0000at://did:plc:ia2a55zrh22gor73moijadrp/app.bsky.feed.post/3mmtm4tzb7225https://bsky.app/profile/baptiste-genest.bsky.social/post/3mmtlceqbxk25When dealing with noisy and/or partial inputs to describe a shape, the standard approach in geometry processing is to reconstruct a clean implicit surface first, and to do geometry on it. But why this one in particular? What about all the other surfaces that would have been coherent with the input?27 May 2026 13:09 +0000at://did:plc:ia2a55zrh22gor73moijadrp/app.bsky.feed.post/3mmtlceqbxk25https://bsky.app/profile/baptiste-genest.bsky.social/post/3m7yozhfq6k2lSuper happy and honored to share that our paper "BSP-OT: Sparse transport plans between discrete measures in log-linear time" won a *Best paper award* at SIGGRAPH Asia 2025! If you are here, come see my presentation about this work Wednesday afternoon! Many thanks to the award committee! [contains quote post or other embedded content]15 Dec 2025 03:31 +0000at://did:plc:ia2a55zrh22gor73moijadrp/app.bsky.feed.post/3m7yozhfq6k2lhttps://bsky.app/profile/baptiste-genest.bsky.social/post/3m256wlpsgs22Computing 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!01 Oct 2025 13:55 +0000at://did:plc:ia2a55zrh22gor73moijadrp/app.bsky.feed.post/3m256wlpsgs22