DeepWiki is awesome! Now you can skip writing GitHub docs 😎 Check mine: deepwiki.com/Vincentqyw/ima…

A Guide to Structureless Visual Localization Vojtech Panek, Qunjie Zhou , Yaqing Ding, Sérgio Agostinho Zuzana Kukelova , Laura Leal-Taixe tl;dr: RoMa beats MAST3r outdoors with 5pt solver, indoors MAST3r is the king. M3Dv2 depth comparable to MAST3r 1/ arxiv.org/abs/2504.17636

This guy literally breaks down everything you need to know about AI in 2 hours

Code dropped: github.com/cvg/mpsfm

New "business card"🚀

Code: github.com/lyp-deeplearni…

FastMap: Revisiting Dense and Scalable Structure from Motion "FASTMAP, a redesigned SfM framework, achieves fast, high-accuracy dense structure from motion. On large scenes with thousands of images, FASTMAP is up to one to two orders of magnitude faster than GLOMAP and COLMAP.

Deepwiki is ready: deepwiki.com/cvg/mpsfm

This has strong implication: it means that the layer from where you extract features is a very important hyperparameter to tune, and it can heavily improve your results And this is true regardless of the task and the model And yet, it's not something people ever tune

DeepSeek just dropped the single best end-to-end paper on large model training. It covers — Software (MLA, training in FP8, DeepEP, LogFMT) — Hardware (Multi-Rail Fat Tree, Ethernet RoCE switches) — Mix (IBGDA, 3FS filesystem) DeepSeek's engineering depth is insane. Must read.

VGGT-SLAM: Dense RGB SLAM Optimized on the SL(4) Manifold Dominic Maggio, Hyungtae Lim, Luca Carlone tl;dr: VGGT->multiple submaps->projective ambiguity->submap alignment->factor graph optimization on the SL(4) manifold (Special Linear, 4 × 4 homography matrix)

Homepage: bagel-ai.org

app: huggingface.co/spaces/wushuan…

MAC-VO: Metrics-Aware Covariance for Learning-based Stereo Visual Odometry TL;DR: learning-based stereo; learned metrics-aware matching uncertainty for dual purposes: selecting keypoint and weighing the residual in pose graph optimization.

Congrats 🚀🚀🚀

Chain-of-Zoom Extreme Super-Resolution via Scale Autoregression and Preference Alignment