I've always appreciated the different experiments on reviewing done by NeurIPS PCs over the years. However the ever increasing verbosity of rebuttals these days does call for ways to reduce length and the 1 page pdf from CVPR does not look that bad or obsolete.

Christian Wolf@chriswolfvision.bsky.social · yesterday

Rebuttals and discussions need to be reduced an extreme minimal length. CVPR style, 1 single PDF page. No discussion. Else, human ACs and reviewers will have a disadvantage compared to the world of agents out there now.

📢 Deadline Extended! We have extended the Student Support Grant application deadline for the ILR+G Workshop @eccv.bsky.social by 2 days. If you haven't applied yet, there's still time! 🗓️ New deadline: August 2, 2026 📝 Apply now: forms.gle/qvqn2LyWXCvN... We look forward to seeing you in Malmö!

Giorgos Kordopatis-Zilos@gkordo.bsky.social · 2w ago

🎓 Student Support Grants are now available! Are you a student working on instance-level recognition and generation? We offer Student Support Grants for participants attending the ILR+G Workshop @eccv.bsky.social 📅 Application deadline: July 31, 2026 📝 Apply here: forms.gle/qvqn2LyWXCvN...

DGSfM: Depth-Guided Scale-Aware Global Structure-from-Motion Sithu Aung,Viktor Kocur,Yaqing Ding, @sattlertorsten.bsky.social Zuzana Kukelova tl;dr: carefully engineered pipeline with dopplegangers++ graph filtering, monodepth, other steps to help GloMAP. Good IMC21 eval arxiv.org/abs/2607.09507

From the org/workshop PC, CMT is miles above OpenReview. Literally, I cannot do anything w/o asking ChatGPT how to do it. Never happened with CMT.

Become a partner of #3DV2027 in Thessaloniki - #CallForSponsors - Your company logo prominently displayed - Complimentary conference registrations - Exhibition space - Many more options… 💡More info: 3dvconf.github.io/2027/partner...

Fable feels differently not only because it is smarter, but also because it fixes small or big issues on its own.

DS-SAC: Density Search for Sample Consensus Suraj Thapa, Muhammad Aminul Islam tl;dr: trimmed LSQ + tree search+bells&whistles= DETERMINISTIC ALG beats RANSAC? They say yes, my Fable re-implementations says "only basic", but I wasn't rigorous (see next) arxiv.org/abs/2607.03972

Glob3R: Global Structure-from-Motion with 3D Foundation Models Junyuan Deng et 9 al. tl;dr: finetune Pi3 matching head in RoMav2 style, use SALAD to sort seqs, do Glomap-like global SfM arxiv.org/abs/2607.09225

AeroMap3D: Anchoring Monocular UAV 6-DoF Localization to Visual-Geometric-Semantic Map Priors Zhiyun Deng, Luis Sentis tl;dr: synthetic data FTW, tiny RoMa is surprisingly good for UAV-Satellite. arxiv.org/abs/2607.14009

CSS-BA: Gate-Guided Column Space Search for Bundle Adjustment Ayano Kaneda, Takafumi Taketomi, Shugo Yamaguchi, Shigeo Morishima tl;dr: solve top-k cameras on low-dim basis -> avoid collapse with weak constraints arxiv.org/abs/2607.15652

Colmap added DEGENSAC option to their RANSAC, merged to main. It helps on plane-dominant scenes and changes nothing in general ones. I should probably retire pydegensac now github.com/colmap/colma...

As a person, who spend some time working on affine feature matches, I do believe, that getting them (if needed) from the dense flow, is the most correct way.

Dmytro Mishkin@ducha-aiki.bsky.social · 2w ago

Rolling Shutter Relative Pose Estimation Made Practical @danielbarath.bsky.social tl;dr: affine correspondences for rolling shutter, because you need 20 PC otherwise. You also can get them from dense matcher like RoMa. P.S. (or we can ignore RS first?) arxiv.org/abs/2606.26863

Argus: Metric Panoramic 3D Reconstruction for Indoor Scenes Xi Li, Linyuan Li, Yan Wu, Tong Rao, Kai Zhang, Xinchen Hui, Cihui Pan tl;dr: panoramas for indoor, and covisibility transformer to select reference frame prior -- important for indoor VGGT. arxiv.org/abs/2606.30047

Rolling Shutter Relative Pose Estimation Made Practical @danielbarath.bsky.social tl;dr: affine correspondences for rolling shutter, because you need 20 PC otherwise. You also can get them from dense matcher like RoMa. P.S. (or we can ignore RS first?) arxiv.org/abs/2606.26863

RANSAC Scoring Done Right James Pritts, Felix Seegräber, Kevin Köser tl;dr: sorts the residuals, then score each inlier/outlier how well they fit the Gaussian for inliers and uniform for outliers. Free param is outlier density. arxiv.org/abs/2606.27385

SeeSE3: Emergence of 3D Space in Vision Features Caroline Chen et 7 al. tl;dr: depth, DINO*, matching features are camera-pose aligned, even better after head ft. Also, scale matters more than 3D?(DINO vs CroCo) I wonder about place recognition models.

SUFLECA: Scaling Up Feature Learning for CAD-to-image Alignment Saad Ejaz, Miguel Fernandez-Cortizas, Javier Civera, Holger Voos, Jose Luis Sanchez-Lopez tl;dr: NOCs (normalized object (surface) coordinates) prediction is good proxy for correspondence train. arxiv.org/abs/2607.15058

#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?...

The Weaver's Song

YouTube video by EchoesOfVelandria - Topic

youtu.be

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.

Giorgos Kordopatis-Zilos@gkordo.bsky.social · 2mo ago

🚨 Call for Papers 🚨 8th Instance-Level Recognition and Generation (ILR+G) Workshop at @eccv.bsky.social 📍 Malmö, Sweden 📅 September 8–9, 2026 🌐 ilr-workshop.github.io/ECCVW2026/ Submission deadline: June 26, 2026 Notification of acceptance: July 24, 2026 #ECCV2026 #ComputerVision

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

SFM-DL - ECCV'26 workshop

ABOUT

sfm-dl.fbk.eu

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.