Jack (in SF) Langerman @jacklangerman.bsky.social · 03/06/2026🎉Workshop On Urban Scene at @cvprconference.bsky.social modeling kicks off in less than 1 hr!!🎉 first up: @niessner.bsky.social ! You don't want to miss this one!! Mile high 3B 010
Jack (in SF) Langerman @jacklangerman.bsky.social · 03/06/2026We have arrived @cvprconference.bsky.social ! it's going to be a great week! 010
Reposted by Jack (in SF) LangermanDmytro Mishkin @ducha-aiki.bsky.social · 02/06/2026Tomorrow is a better day...because USM3D workshop at @cvprconference.bsky.social is happening? 6 keynote speakers. 2 challenges which I am finally not ashamed of? Debatable. Is it our final competition at HuggingFace? The only way to figure out is to attend. Mile High 3B. #CVPR2026 091
Jack (in SF) Langerman @jacklangerman.bsky.social · 03/06/2026🌟🌟 Start @CVPR off right with the Workshop on Urban Scene Modeling! (Tues Jun 2) @ Mile High 3B! 🌟 - Great lineup of key note speakers! - Challenge winners will be announced! - Selected paper talks! 3D/4D reconstruction+generation+understanding applied to the built world 100
Reposted by Jack (in SF) LangermanDmytro Mishkin @ducha-aiki.bsky.social · 18/03/2026Sad, but important announcement -- one of our challenges -- Building3D lost their sponsor in the very last moment. So new prize fund is $14k: - $12k for S23DR 2026 - $2k for Building3D 2026. We apologise for this situation, and hope that you still decide to participate. #CVPR2026 061
Jack (in SF) Langerman @jacklangerman.bsky.social · 12/03/2026ive taken to intentionally using endashes so i still get to use dash-based parentheticals, but astute observers will know it was "made by humans™" 230
Reposted by Jack (in SF) LangermanDmytro Mishkin @ducha-aiki.bsky.social · 04/03/2026Submit your paper of structured reconstruction -- CAD, semantic, wireframe, city monitoring, etc., to USM3D 2026! cmt3.research.microsoft.com/USM2026 Deadline: March 24, 2026. @cvprconference.bsky.social #CVPR2026 #USM3D2026 #USM3D 053
Jack (in SF) Langerman @jacklangerman.bsky.social · 17/02/2026at least the pytorch team triaged this one (that i found) and it got fixed github.com/pytorch/pyto... it really should be a priority though for sure - easy win imogithub.comMPS compile off by 2x with broadcast and sum · Issue #170360 · pytorch/pytorch🐛 Describe the bug Hey team, Looks like there is is a correctness issue with torch.compile when mode="default" on MPS for this simple broadcast and sum function. I'm seeing the output almost exactl... 010
Jack (in SF) Langerman @jacklangerman.bsky.social · 04/02/2026This was joint work with my wonderful colleagues Denis Rozumny, Yuzhong Huang, and @ducha-aiki.bsky.social and was presented at @iccv.bsky.social last year. bsky.app/profile/jack... 010
Jack (in SF) Langerman @jacklangerman.bsky.social · 04/02/2026🧵A bit last minute but I'm giving a talk tonight in SF about my work Explaining Human Preferences via Metrics for Structured 3D Reconstruction (ICCV Highlight) Quite near transamerica pyramid. Event is already full, but comment or shoot me a DM and we'll find a spot for you! 110
Reposted by Jack (in SF) LangermanDmytro Mishkin @ducha-aiki.bsky.social · 24/10/2025Come to our poster this afternoon starting at 2:30 (#214) 031
Jack (in SF) Langerman @jacklangerman.bsky.social · 10/10/2025What is the right metric to measure performance in structured 3D reconstruction? 🎬 Just posted the video for our @iccv.bsky.social #ICCV2025 highlight paper! 🎉 @jacklangerman.bsky.social, Denys Rozumnyi, Yuzhong Huang, @ducha-aiki.bsky.social 🌄 iccv.thecvf.com/virtual/2025...youtu.be[ICCV 2025] Explaining Human Preferences via Metrics for Structured Reconstruction - ICCV HighlightYouTube video by Jack Langerman 021
Jack (in SF) Langerman @jacklangerman.bsky.social · 22/09/2025what makes a good paper video? what are some of the best you have seen? cc: @davidpicard.bsky.social @davidbau.bsky.social @eugenevinitsky.bsky.social @chriswolfvision.bsky.social @nsaphra.bsky.social @ducha-aiki.bsky.social @parskatt.bsky.social @martin-r-oswald.bsky.social @ericzzj.bsky.social 221
Jack (in SF) Langerman @jacklangerman.bsky.social · 13/09/2025iiim8uhl j ij9j b knmnjj mlpnnom l 8kc6 Kimberly.C mbkln9j8 . ojlioij7o.h ihlokMO Lip l 9 b nl d9 100
Jack (in SF) Langerman @jacklangerman.bsky.social · 25/06/2025Very proud that our paper has been accepted to @iccv.bsky.social !!! See you in Hawaii! bsky.app/profile/duch... 061
Reposted by Jack (in SF) LangermanZhenjun Zhao @ericzzj.bsky.social · 09/05/2025🎉 Thrilled to share our CVPR 2025 Award Candidate & Oral paper: 🔹 GlobustVP Convex Relaxation for Robust Vanishing Point Estimation in Manhattan World 🧱 Global optimality 💥 Tolerates up to 70% outliers ⚡ Fast runtime 📄 Paper: arxiv.org/abs/2505.04788 💻 Code: github.com/WU-CVGL/GlobustVP 1/ 1315
Reposted by Jack (in SF) Langerman#ICCV2025 @iccv.bsky.social · 09/05/2025#ICCV2025 reviews are out and being sent via email to authors! They will also be available on OpenReview later. 11,152 active submissions all have at least 3 reviews. Authors have the opportunity to submit a rebuttal by May 16 2025 11:59 PM HST. 1198
Jack (in SF) Langerman @jacklangerman.bsky.social · 07/05/2025Please reshare, tell you friends, give it a try, and don't hesitate to reach out to the team with any questions! bsky.app/profile/jack... 010
Jack (in SF) Langerman @jacklangerman.bsky.social · 07/05/2025🌍 Workshop + other challenges at: usm3d.github.io Let’s make SfM more structured. Good luck, teams! (7/7)usm3d.github.io2nd Workshop on Urban Scene Modeling: Where Vision Meets Photogrammetry and GraphicsCVPR 2025 Workshop on Urban Scene Modeling: Where Vision Meets Photogrammetry and Graphics. 111
Jack (in SF) Langerman @jacklangerman.bsky.social · 07/05/2025📌 Want a clear reference? Check out the example baseline submission: huggingface.co/usm3d/handcr... (6/7)huggingface.cousm3d/handcrafted_submission_2025 · Hugging FaceWe’re on a journey to advance and democratize artificial intelligence through open source and open science. 100
Jack (in SF) Langerman @jacklangerman.bsky.social · 07/05/2025📊 NEW Evaluation Metric Say goodbye to WED. Hello to a human-aligned combo of: • Vertex F1 • Edge IoU Backed by this paper: 👉 arxiv.org/abs/2503.08208arxiv.orgExplaining Human Preferences via Metrics for Structured 3D Reconstruction"What cannot be measured cannot be improved" while likely never uttered by Lord Kelvin, summarizes effectively the purpose of this work. This paper presents a detailed evaluation of automated metrics ... 100
Jack (in SF) Langerman @jacklangerman.bsky.social · 07/05/2025⚙️ All helper functions + data loaders: pip install hoho2025 🛠️ Codebase: github.com/s23dr/hoho2025 (4/7)github.comGitHub - s23dr/hoho2025: Tools and utilities for the S23DR-2025 competition and HoHo25k DatasetTools and utilities for the S23DR-2025 competition and HoHo25k Dataset - s23dr/hoho2025 100
Jack (in SF) Langerman @jacklangerman.bsky.social · 07/05/2025📦 New dataset: hoho25k — 5x bigger than 2024! • 25,000 scenes • 8 images per scene • 200,000 total images 📥 Get it here: huggingface.co/datasets/usm... (3/7)huggingface.cousm3d/hoho25k · Datasets at Hugging FaceWe’re on a journey to advance and democratize artificial intelligence through open source and open science. 110
Jack (in SF) Langerman @jacklangerman.bsky.social · 07/05/2025🧠 The task: Turn multiview inputs (semantic segs, monocular depth) + SfM outputs (point clouds + camera poses) into sparse, geometrically accurate wireframes. AKA: “More Structured Structure-from-Motion” 😎 Challenge page 👉 huggingface.co/spaces/usm3d... (2/7)huggingface.coS23DR2025 - a Hugging Face Space by usm3dThis application allows you to view competition details, manage your submissions, and check leaderboards. You can access rules, dataset information, and view your own submissions and scores. 110
Jack (in SF) Langerman @jacklangerman.bsky.social · 07/05/2025🚨 Just one month left to submit your solutions for The Structured Semantic 3D Reconstruction (S23DR-2025) Challenge!!! It is not too late to join! Comp on @hf.co, part of the Workshop on Urban Scene Modeling at @cvprconference.bsky.social 2025 🔥$25,000 prize pool. Deadline: June 5, 2025. 🧵 (1/7) 154
Jack (in SF) Langerman @jacklangerman.bsky.social · 30/04/2025hey @cloneofsimo.bsky.social (or anyone else) did you ever try LoRA+ style LR split between down/up LoRA projection matrices on diffusion models? 010
Reposted by Jack (in SF) LangermanAaron Mueller @amuuueller.bsky.social · 23/04/2025Lots of progress in mech interp (MI) lately! But how can we measure when new mech interp methods yield real improvements over prior work? We propose 😎 𝗠𝗜𝗕: a 𝗠echanistic 𝗜nterpretability 𝗕enchmark! 15115
Jack (in SF) Langerman @jacklangerman.bsky.social · 27/03/2025When I was working at Bell Labs in Murray Hill we used to specifically go to 3rd floor for coffee because the couch in that particular coffee room was (supposedly) Dennis Ritchie's. was also a room with a whiteboard that may have been where Shannon did a lot of the info theory work. good times 020
Jack (in SF) Langerman @jacklangerman.bsky.social · 18/03/2025do you think this is a matter of semantic disipline or genuine beliefs about qualia? ie "with certain inputs models can produce outputs that contain patterns typically associated with anxiety in humans. asking them to perform mindfulness exercises can mitigate this behavior" vs "they feel anxious"? 020
Reposted by Jack (in SF) LangermanDmytro Mishkin @ducha-aiki.bsky.social · 17/03/20251 week to USM3D deadline 063
Jack (in SF) Langerman @jacklangerman.bsky.social · 15/03/2025more blending between rings i think (~long range dependence) 110
Reposted by Jack (in SF) LangermanDmytro Mishkin @ducha-aiki.bsky.social · 12/03/2025Explaining Human Preferences via Metrics for Structured 3D Reconstruction @jacklangerman.bsky.social Denys Rozumnyi, Yuzhong Huang, @ducha-aiki.bsky.social tl;dr: we asked 3D modelers to rank wireframe reconstructions & compared it to ranking by metrics. Observations🧵 1/ arxiv.org/abs/2503.08208 12810
Reposted by Jack (in SF) LangermanDmytro Mishkin @ducha-aiki.bsky.social · 10/03/20252nd Building3D #CVPR2025 challenge at #USM3D workshop is open! Task: point cloud to wireframe. Prize pool: $10k Competition deadline: May 25 2025. Website: huggingface.co/spaces/Build... @cvprconference.bsky.social 0124
Reposted by Jack (in SF) LangermanJohan Edstedt @parskatt.bsky.social · 10/03/2025We made a new keypoint detector named DaD, paper isn't up yet, but code and weights are: github.com/Parskatt/dad 7448
Reposted by Jack (in SF) LangermanDmytro Mishkin @ducha-aiki.bsky.social · 10/03/2025Those, who work in structured (images/pcl to CAD) reconstruction - USM3D #CVPR2025 workshop submissions are open. Deadline: March 24 2025 Both full papers (8 pages) and extended abstracts (4 pages) are OK usm3d.github.io #USM3D @cvprconference.bsky.social @jacklangerman.bsky.social 073
Reposted by Jack (in SF) LangermanDmytro Mishkin @ducha-aiki.bsky.social · 09/03/2025We have extended the deadline for paper submission for Image Matching Workshop. Anything image matching or 3D reconstruction related is welcomed Now it is March 17 @cvprconference.bsky.social #CVPR2025 image-matching-workshop.github.io 1178
Jack (in SF) Langerman @jacklangerman.bsky.social · 06/03/2025github.com/pytorch/pyto...github.comMPS operator coverage tracking issue (2.6+ version) · Issue #141287 · pytorch/pytorch🐛 Describe the bug This issue is to have a centralized place to list and track work on adding support to new ops for the MPS backend. PyTorch MPS Ops Project : Project to track all the ops for MPS ... 130
Jack (in SF) Langerman @jacklangerman.bsky.social · 06/03/2025yeah `PYTORCH_ENABLE_MPS_FALLBACK=1 python myscript.py` did it. 010
Jack (in SF) Langerman @jacklangerman.bsky.social · 31/01/2025what is the best intro I can send people who suddenly want to know "what is AI?" and "what is a transformer based model?" but arent really going to invest tons of time and maybe dont have any math background? my default is 3b1b, but does anyone have another suggestion? 000
Reposted by Jack (in SF) LangermanDmytro Mishkin @ducha-aiki.bsky.social · 02/01/2025Image matching and ChatGPT - new post in the wide baseline stereo blog. tl;dr: it is good, even feels like human, but not perfect. ducha-aiki.github.io/wide-baselin...ducha-aiki.github.ioChatGPT and Image Matching – Wide baseline stereo meets deep learningAre we done yet? 2348
Jack (in SF) Langerman @jacklangerman.bsky.social · 29/12/2024ive said for a long time this is one of the things I love most about ML: get to play in everyones' playgrounds :-) 020
Jack (in SF) Langerman @jacklangerman.bsky.social · 21/12/2024Yeah, definitely a healthy amount of uncertainty here, but somehow feels related to bsky.app/profile/jack... Maybe we just don't have the right optimizer or something, so we need the easy to train big model to "simplify" for the small (harder to optimize) model, but for sure a bit of a mystery 120
