Here's a quick thread about how the procedural material is set up to enhance the details of the editable terrain in Island Architect, without baking any textures. Still details I want to add, but I'm happy with the foundation! Here's the final result and the diffuse texture: #gamedev #indiegame

Anthropic: "Mythos is too dangerous". Claude Code, shortly after: "API Error: 500 {"type":"error","error":{"type":"api_error","message":"Internal server error"}". Has machine rebellion already started?

Is it a csv, when the file has single column and no separator?

With LLMs and coding agents "doing things right" becomes way easier than before. Therefore doing things in a bad way is way less excusable.

🚨 arxiv.org/abs/2604.06129 PoM: A Linear-Time Replacement for Attention with the Polynomial Mixer This paper is the result of doing a lab-wide hackathon on an idea I've had for some time. Probably the paper with the highest number of authors I've ever done. It's a CVPR Findings 26. Thread 🧵👇

LoMa: Local Feature Matching Revisited @davnords.bsky.social @parskatt.bsky.social , @bokmangeorg.bsky.social et 6 al tl;dr: if you train DeDoDe+LightGlue on VGGT-scale data, it helps a LOT. New IMC2022 sota arxiv.org/abs/2604.04931

A problem with current image matching benchmarks is that they are starting to saturate, and another that they are not diverse enough. HardMatch solves this (see Fig 1 b), by being way harder than existing benchmarks, while being much more diverse.

There is a joke that when you are writing to academic, it is equally probable to get an answer in 2 minutes and 2 years. It is not a joke - I have just closed 6 years old GH issue on pydegensac.

Related read: ergosphere.blog/posts/the-ma...

Is there anything better than good old mask2former for segmentation? P.S. I treat OneFormer and OnePiece as mask2former with minor changes

I am definitely going to first try my new model!

New "AI" gmail search is a disaster. When I search for "nostrification", I don't want to see emails with "recognition". When I search for "compensation" -- in a tech sense, I don't way to see tons of email with "pay". Especially with "pay attention".

Excited to announce our latest (submitted to) SIGBOVIK 2026 @harryqbovik.bsky.social paper: "SchmidhubAI: Accurate Historical Paper Attribution". We built an AI system that, given any modern AI paper, automatically determines which of its ideas were already published by Jürgen Schmidhuber.

My 2 cents on the impact of Agentic AI on peer review: it's not looking good. medium.com/@chriswolfvi...

Sky2Ground: A Benchmark for Site Modeling under Varying Altitude Zengyan Wang, Sirshapan Mitra, Rajat Modi, Grace Lim, Yogesh Rawat tl;dr: Dataset + how to adapt VGGT for satellite + ground imagery. Of course, the method is named SkyNet. arxiv.org/abs/2603.13740

VFM-Loc: Zero-Shot Cross-View Geo-Localization via Aligning Discriminative Visual Hierarchies Jun Lu, Zehao Sang, Haoqi Wei, Xiangyun Liu, Kun Zhu, Haitao Guo, Zhihui Gong, Lei Ding tl;dr: drone (high height) vs satellite with DINOv3+RMAC/GeM. arxiv.org/abs/2603.13855

UE5-Forest: A Photorealistic Synthetic Stereo Dataset for UAV Forestry Depth Estimation Yida Lin, Bing Xue, Mengjie Zhang, Sam Schofield, Richard Green tl;dr: the dataset is small and not available online, but check the paper if you plan to do similar arxiv.org/abs/2603.15304

TreeON: Reconstructing 3D Tree Point Clouds from Orthophotos and Heightmaps Angeliki Grammatikaki Johannes Eschner Pedro Hermosilla, Oscar Argudo, Manuela Waldner tl;dr: in title. arxiv.org/abs/2603.10996

Copying here @damienteney.bsky.social tweet. I cannot stop think about this now.

Once I told Claude to think like John Carmack, it was able to speed up code 10x.

Jensen's inequality

Benchmarking Efficient & Effective Camera Pose Estimation Strategies for Novel View Synthesis Jhacson Meza, @martin-r-oswald.bsky.social, @sattlertorsten.bsky.social arxiv.org/abs/2603.20428

🚀 New task: Instance-level Composed Image Retrieval 🔎 Given a [query image] + [query text], retrieve the particular object after the change — not just any similar object. 🛢 New dataset on HF: i-CIR huggingface.co/datasets/bil... 📄 Project page: vrg.fel.cvut.cz/icir/

Make sure to come by our poster right now if you are at #3DV2026!

Sad, 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

We need more bold people like this. All conferences should experiment, and take action for rules violation.

MLE seeking team About you: - Your organization is looking to do something new in ML. - You want to build a new team, a new product direction, a new capability. - You care more about what gets accomplished than about the tools used to do it. - You foster accountability and high trust teams.

Two #CVPR2026 competitions are live: S23DR 2026 and BuildingWorld 2026! Task: reconstruct house roof wireframes from point clouds and segmentations: Total prize fund: $22k Deadline: end of May 2026 1) huggingface.co/spaces/usm3d... 2) huggingface.co/spaces/Build... @cvprconference.bsky.social

PixARMesh: Autoregressive Mesh-Native Single-View Scene Reconstruction Xiang Zhang, Sohyun Yoo, Hongrui Wu, Chuan Li, Jianwen Xie, Zhuowen Tu tl;dr: image, depth, instance seg tokens into auto-regressive decoder to get pose and mesh.Pose as bbox vertices to share with mesh arxiv.org/abs/2603.05888

Great reason to clean your monitor :)

Topologically Stable Hough Transform Stefan Huber, Kristóf Huszár, Michael Kerber, Martin Uray tl;dr: replace voting with KDE in Hough transform. Then do kind of MSERs, but not quite, to get the global and stable peaks. arxiv.org/abs/2603.08245 github.com/JRC-ISIA/Top...

Wow! Never thought I would see it on the internet ever again. I've been using this picture as an avatar for at least 20 years. Glad it's available in higher resolution than it used to.

You cannot make this thing up

VGGT-Det: Mining VGGT Internal Priors for Sensor-Geometry-Free Multi-View Indoor 3D Object Detection Yang Cao, Feize Wu, Dave Zhenyu Chen, Yingji Zhong, Lanqing Hong, Dan Xu tl;dr: VGGT feature are not that great for detection unless you bell&whistle. arxiv.org/abs/2603.00912

Multimodal Large Language Models as Image Classifiers Nikita Kisel, Illia Volkov @klara-cz.bsky.social Jiri Matas tl;dr: if you evaluate good (chatGPT) model on a dirty (ImageNet) test set, it is bad. Yes, ImageNet test is bad nowadays. +insights from labeling. arxiv.org/abs/2603.065...

🧵For the last seven years, I kept re-implementing the same pattern: A parallel map loop that divides the work among several processes or threads. My very first attempts were built on Python’s standard tools, e.g., multiprocessing.map... ↩️

To study this, we introduce ReGT, a new multilabel reannotation of 625 ImageNet classes that corrects many of these issues. When evaluated on the cleaned labels, multimodal LLMs improve by up to +10.8% accuracy, substantially narrowing the gap with supervised vision models. 📈

Let me introduce our new paper: Multimodal Large Language Models as Image Classifiers ❓ Multimodal LLMs are increasingly used for visual tasks, but evaluating their image classification ability has produced conflicting conclusions. Link: arxiv.org/html/2603.06...

I've made a SatAst, a small collection of hand-annotated satellite to astronaut image correspondences, public on github: github.com/georg-bn/sat.... This benchmark is part of the RoMa v2 paper, see Johan's thread below. bsky.app/profile/pars...

"Authors should not use negative v-spaces to change the template layout." The template layout:

Utonia: Toward One Encoder for All Point Clouds Yujia Zhang, Xiaoyang Wu, Yunhan Yang, Xianzhe Fan, Han Li, Yuechen Zhang, Zehao Huang, Naiyan Wang, Hengshuang Zhao tl;dr: PointTransformerv3 pretrained on tons of different data arxiv.org/abs/2603.03283

ZipMap: Linear-Time Stateful 3D Reconstruction with Test-Time Training @haian-jin.bsky.social Rundi Wu, Tianyuan Zhang, Ruiqi Gao, @jonbarron.bsky.social @snavely.bsky.social @holynski.bsky.social tl;dr: more test-time-training for getting scene latent. arxiv.org/abs/2603.04385

DAGE: Dual-Stream Architecture for Efficient and Fine-Grained Geometry Estimation Tuan Duc Ngo, Jiahui Huang, Seoung Wug Oh, Kevin Blackburn-Matzen, Evangelos Kalogerakis, Chuang Gan, Joon-Young Lee tl;dr: low-res multivew (Pi3-distilled) + highres single view( MoGe2 ft) arxiv.org/abs/2603.03744

NOVA3R: Non-pixel-aligned Visual Transformer for Amodal 3D Reconstruction Weirong Chen, Chuanxia Zheng, Ganlin Zhang, Andrea Vedaldi, Daniel Cremers tl;dr: let VGGT output latents -> decode point cloud. arxiv.org/abs/2603.04179

Dark3R: Learning Structure from Motion in the Dark Andrew Y Guo, Anagh Malik, SaiKiran Tedla, Yutong Dai, Yiqian Qin, Zach Salehe, Benjamin Attal, Sotiris Nousias, Kyros Kutulakos, David B. Lindell tl;dr: LoRa for MASt3R to make it work on low-light. arxiv.org/abs/2603.05330 #CVPR2026

NOVA3R: Non-pixel-aligned Visual Transformer for Amodal 3D Reconstruction Weirong Chen, @chuanxiaz.bsky.social, @ganlinzhang.xyz, Andrea Vedaldi, @dcremers.bsky.social tl;dr: TripoSG+VGGT layout issue with tables? arxiv.org/abs/2603.04179

Submit 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

#CVPR2026 One more week to submit your work to the Embedded Vision Workshop @ CVPR! @cvprconference.bsky.social (new deadline: March 11) Info at: embeddedvisionworkshop.wordpress.com

AI writing is like store-bought cake. It might be perfectly fine, maybe even as good as something you could make yourself, but it’s weird to give it to someone and say it’s homemade