#3DV2027 #CallForPapers 3dvconf.github.io/2027/call-fo...

Interested in neural SLAM or online semantic scene understanding? Consider submitting to our #ECCV2026 NeuSLAM workshop! Accepted or rejected, your recent submission can gain more visibility through our proceedings or nectar track! More info: sites.google.com/view/neuslam...

Memorization is not expertise but there’s not really expertise without memorization either

We are aware of the strong interest in the #ECCV2026 camera-ready instructions. Here are the key details while you wait for the Meteor submission site invite: 1/3

For your Embodied AI task you want a recurrent model with constant complexity per step, but you don't want to lose the power of transformers (which store the full obs history and attend to it)? Do not despair, we have your back. We distill transformers into recurrent transformers 1/8

OK, here is the first AI song I genuinely enjoy now. That’s the end? youtu.be/pZt9TEwZZZY?...

Question on the squish/pad RoMa v1, v2, LoMa squish and happy. On the other hand, VGGT, *3R family use complex pipeline: N possible sizes, resize-to-nearest+0 pad. Have you tried the opposite approach? @jianyuanwang.bsky.social @parskatt.bsky.social @davnords.bsky.social @vincentleroy.bsky.social

The instance level recognition and generation workshop will be back at @eccv.bsky.social 2026. With excellent keynote speakers, a call for papers, and travel grants for students.

Have a nice SfM/Image matching idea/paper? Looking for a home for ECCV rejected paper? Want to visit Sweden in September? Submit to #ECCV2026 Workshop SfM in Age of Deep Learning Deadline: July 10, 2026 Decision: July 31, 2026 sfm-dl.fbk.eu @eccv.bsky.social

We have created a static website for all S23DR competitions since 2024. Write-ups, leaderboards, links to the code -- all here s23dr.github.io

S23DR 2026 Winning Solution Jan Skvrna Miroslav Purkrabek, @lukasneumann.bsky.social tl;dr: Denoising flow-matching for point cloud to wireframe. arxiv.org/abs/2606.06695

Edge Prediction for Roof Wireframe Reconstruction with Transformers Gustav Hanning, Ludvig Dillén, @jastermark.bsky.social Johanna Lidholm, Viktor Larsson tl;dr: DETR on point cloud for predicting wireframe. 2nd place in S23DR-2026 challenge. arxiv.org/abs/2606.02406

I don't care much about the CVPR Best Paper Award criteria. Usually it means nothing, as least much less than Test of Time award. It never mattered before, it will never matter in the future. I will change my mind, when the "Best Paper" will also get test of time award.

It's now, wake up! ☕🥐

We solved the technical issues and now @jianyuanwang.bsky.social keynote "From Reconstruction to Perception" -- VGGT father Room 504 #CVPR2026

Nikhil Keetha: "YLC cake is broken for spatial models" "my cake is not layered, it is mixed" @cvprconference.bsky.social #CVPR2026

Get ready to Image Matching Workshop 2026 today at @cvprconference.bsky.social Room: 504 13:00 local time 3 star speakers: ⭐ @pesarlin.bsky.social ⭐ Nikhil Keetha ⭐ @jianyuanwang.bsky.social ✅10 papers ✅IMC2026 story #CVPR2026

Get ready to Image Matching Workshop 2026 today at @cvprconference.bsky.social Room: 504 13:00 local time 3 star speakers: ⭐ @pesarlin.bsky.social ⭐ Nikhil Keetha ⭐ @jianyuanwang.bsky.social ✅10 papers ✅IMC2026 story #CVPR2026

Florent Lafarge keynote in 5 min at #USM3D2026 Mile High 3B. #CVPR2026

The lack of analytics (by default) & engagement here hinders adoption, no? Any workaround post analytics? I'm just live tweeting in X in case you're there #CVPR2026

BuildingWorld dataset and competition #USM3D2026 43 cities

𝑹𝒆𝒕𝒓𝒊𝒆𝒗𝒆 𝒂𝒏𝒅 𝑺𝒆𝒈𝒎𝒆𝒏𝒕 (𝑹𝑵𝑺) at CVPR 2026 — selected as a 🌟 𝑯𝒊𝒈𝒉𝒍𝒊𝒈𝒉𝒕 🌟 (top ~4% of submissions)! 𝗪𝗵𝗲𝗿𝗲 𝘁𝗼 𝗳𝗶𝗻𝗱 𝘂𝘀 𝗮𝘁 𝗖𝗩𝗣𝗥 𝟮𝟬𝟮𝟲 (𝗗𝗲𝗻𝘃𝗲𝗿): 🔹 𝑴𝒂𝒊𝒏 𝑪𝒐𝒏𝒇𝒆𝒓𝒆𝒏𝒄𝒆 Poster Session 4 (#578) — June 6, 16:45–18:45 🔹 𝑾𝒉𝒂𝒕 𝒊𝒔 𝑵𝒆𝒙𝒕 𝒊𝒏 𝑴𝒖𝒍𝒕𝒊𝒎𝒐𝒅𝒂𝒍 𝑭𝒐𝒖𝒏𝒅𝒂𝒕𝒊𝒐𝒏 𝑴𝒐𝒅𝒆𝒍𝒔? Workshop — June 3, 14:30–16:00

3D reconstruction history from @niessner.bsky.social #USM3D2026

Tomorrow 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

Feels slow at times, but I’ve certainly made some progress on my relaxing building game 🌳🏡Island Architect🏡🌳! #screenshotsaturday #indiegame #gamedev #indiegamedev

All those are transformers, btw.

Image matching since 2020: 2020: @pesarlin.bsky.social SuperGlue 2023: @vincentleroy.bsky.social DUSt3R 2024: @parskatt.bsky.social RoMa 2025: @jianyuanwang.bsky.social VGGT 2026: @davnords.bsky.social"hold my beer" scales LightGlue) 2026: @jianyuanwang.bsky.social "no, hold MY beer"(scales VGGT)

VGGT-Ω @jianyuanwang.bsky.social et 9 al. tl;dr: 1) simplicity to scale compute(no pcl head, no matching head, but losses kept, no rays for intrinsics, less DPT) 2) Complex video SfM/filtering procedure to scale data. 3) works great on IMC-2025(not in paper, I tested) arxiv.org/abs/2605.15195

Kocur, Aung, and Flood, et al., "Depth2Pose: A Pose-Based Benchmark for Monocular Depth Estimation without Ground-Truth Depth" You can use camera poses to benchmark depth estimators as well. Each model seem ot have their own bias towards which kind of scene they are best at.

VGGT-Omega paper is so pleasant to read. Unlike most of papers which I just skim through, I enjoy this one and read every sentence. @jianyuanwang.bsky.social et al - great job!

The more I experiment, the more I believe that "mid layers know more geometry" etc is mostly an artifact of probing frozen models. Same about the pooling. GeM is great for frozen backbone. If you can train, use attention.

📢📢📢 While you are all engaging in lively discussions during the rebuttal phase, we want to announce that the #ECCV2026 conference dates have been updated! It has been decided that the conference will end on Saturday the 12th September. Please have this in mind when making travel plans

Apparently #CVPR2026 had an outstanding AC initiative. Well done, @davidpicard.eurosky.social!

We are implementing a similar policy at @arxiv.bsky.social. If there is incontrovertible evidence of LLM slop in a paper, this means the authors did not take the time to read the LLM output and we can't trust anything else in the paper. Penalty is 1 year ban from arXiv followed by...

Claude: "Where I'd put my money" Me: "Bro, you don't have any money"

The quantification trap, or how well-intentioned attempts at intersubjectivity breed institutionalized postmodern optimization. (Part 2 in my series on what I wish I had more clearly articulated in The Irrational Decision.)

Claude, after his clone was working on a feature for an hour: "The branch is now 3 weeks behind on a pipeline-level feature"

How dare you exploit academic labor; that’s our job!

Agent: Let’s start slow and keep it simple. Me: Great. Now make it robust. Agent: Absolutely. I’ll preserve backward compatibility with all the mistakes we made 30 seconds ago.

There's an increasingly popular argument saying that for 3D point clouds with rotational ambiguity (e.g. molecular data) standard transformers are preferable over equivariant models for compute reasons. With “Platonic Transformers” we show why this doesn't have to be true. arxiv.org/abs/2510.03511

NeurIPS'26 (abs): DL today, good luck (2h)! BMVC'26 (abs): 16 days. CORL'26 (abs): 18 days. CORL'26 (paper): 21 days. BMVC'26 (paper): 23 days.

Yes.

Match-Any-Events: Zero-Shot Motion-Robust Feature Matching Across Wide Baselines for Event Cameras Ruijun Zhang, Hang Su, Kostas Daniilidis, Ziyun Wang tl;dr: spatio-temporal transformer aggregation -> (ViT+DPT) for matching. New Event-MegaDepth dataset arxiv.org/abs/2604.18744

Triangulation of Points Constrained to a Plane @petrhruby.bsky.social Elima Shehu tl;dr: use DAv2 to get planar regions -> apply new solvers for those to get more precise triangulation. arxiv.org/abs/2604.27246

If you are going to use LLM to do the paper review (don’t), at least do it properly: - use Deep research mode - ask Claude/ChatGPT write a detailed deep research prompt first.

OK let’s see if people are better than Claude in literature review. Is there a paper, comparing output representation vs supervision? E.g. there is a consensus that predicting rays is better, than K matrix directly. But what is better- supervising K with rays, or rays with K?

Non-Minimal Sampling and Consensus for Prohibitively Large Datasets Seong Hun Lee, Patrick Vandewalle, Javier Civera tl;dr: you split (huuuge) tentatives list, and run RANSAC of each. How do you combine winning hypothesis? Truncated Lp cost seems the best. arxiv.org/abs/2604.22518

ArchSym: Detecting 3D-Grounded Architectural Symmetries in the Wild Hanyu Chen, @ruojin.bsky.social Steve Marschner, @snavely.bsky.social tl;dr: dataset + finetuned VGGT for symmetry detection arxiv.org/abs/2604.22202

Long-tail Internet photo reconstruction Yuan Li, Yuanbo Xiangli, Hadar Averbuch-Elor, @snavely.bsky.social @ruojin.bsky.social tl;dr: Cleaned-up version of MegaScenes + MoGe2-based depth refinement. +sparse sampler for emulate hard scenes. arxiv.org/abs/2604.22714