🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

Absolutely elated to share that my paper with Prof. Andrew Zisserman and Joao Henriques Visual Geometry Group (VGG) about "incorporating multi-view geometric priors into transformers" has been accepted to CVPR 2023!! 😇🥳 #CVPR2023 #cvpr #computervision #3D

Congratulations to Edgar who passed his PhD viva today, and thanks to examiners Tolga Birdal and José María Montiel! A reminder of Edgar's iMAP, a landmark as the first real-time neural field SLAM system from #ICCV2021.

Excited to present our #ICASSP2024 work ‘A Sound Approach: Using LLMs to Generate Audio Descriptions for Egocentric Text-Audio Retrieval’ which introduces egocentric text-audio retrieval benchmarks. Poster: MLSP-P8.8 (Apr 16th, 16:30-18:30) Paper: arxiv.org/abs/2402.19106

Reconstructing an 𝘂𝗻𝗯𝗼𝘂𝗻𝗱𝗲𝗱 𝘀𝗰𝗲𝗻𝗲 𝗳𝗿𝗼𝗺 𝗮 𝗳𝗲𝘄 𝗼𝘂𝘁𝘄𝗮𝗿𝗱-𝗳𝗮𝗰𝗶𝗻𝗴 𝗶𝗺𝗮𝗴𝗲𝘀 is tough. We present 6Img-to-3D, an efficient, scalable single-shot image to 3D reconstruction method.

github.com/TimSong412/2D-… Our wonderful master students TimSong and his classmates re-implemented the 2DGS! a bold step towards surfels and geometry reconstruction from Gaussian Splatting!

Thanks AK for featuring our work! We present a depth completion network that allows generating geometrically coherent 3D scenes. See the project page research.paulengstler.com/invisible-stit… for more results or try it yourself huggingface.co/spaces/pauleng… on HF🤗

Meta releases VGGSfM Visual Geometry Grounded Deep Structure From Motion Structure-from-motion (SfM) is a long-standing problem in the computer vision community, which aims to reconstruct the camera poses and 3D structure of a scene from a set of unconstrained 2D images.

Exciting research at #ECCV2024! 🔍 Our paper "Dataset Enhancement with Instance-Level Augmentations" introduces generative image augmentations. Interested? Come learn more: 🗣️ Talk today @ 14:20 - Auditorium 🖼️ Poster #145 @ 16:30 We'd love to discuss it with you!

Our paper “Dissecting temporal understanding in text-to-audio retrieval” is being presented today ACM Multimedia. Poster: P189 - Session 3 Paper: arxiv.org/abs/2409.00851 Code: github.com/oncescuandreea…

Hi! If you found the gsplat library (docs.gsplat.studio) useful, we wrote a whitepaper with benchmarking, conventions, derivations, and new features (great effort led by Matias Turkulainen & Ruilong Li 🙌). arxiv.org/abs/2409.06765

Joao Henriques (joao.science) and I are hiring a fully funded PhD student (UK/international) for the FAIR-Oxford program. The student will spend 50% of their time University of Oxford and 50% AI at Meta (FAIR), while completing a DPhil (Oxford PhD). Deadline: 2nd of Dec AOE!!

Check out our work on feed-forward (in a flash) 3D scene reconstruction! Our method transforms a monocular depth estimation network into a generalisable single-view 3D reconstructor. Thanks to Stan Szymanowicz, Chuanxia Zheng, Dylan Campbell, J. Henriques, @chrirupp & A. Vedaldi!

Hi there, 🎉 We are thrilled to introduce Stable Virtual Camera, a generalist diffusion model designed to address the exciting challenge of Novel View Synthesis (NVS). With just one or a few images, it allows you to create a smooth trajectory video from any viewpoint you desire.

We are presenting Dual Point Maps as a #CVPR highlight tomorrow! Learn about our novel, data-efficient representation for 3D/4D deformable objects—an alternative to classical template shape models. 📍🕑 ExHall D, Poster #100, afternoon session 🌍dualpm.github.io

Many Congratulations to Jianyuan#CVPR2025 2025, Minghao Chen, Nikita Karaev, Andrea Vedaldi, Christian Rupprecht and David Novotny for winning the Best Paper Award @CVPR for "VGGT: Visual Geometry Grounded Transformer" 🥇🎉 🙌🙌 #CVPR2025!!!!!!

Excited to share VMem: a novel memory mechanism for consistent video scene generation 🎞️✨ VMem evolves its understanding of scene geometry to retrieve the most relevant past frames, enabling long-term consistency 🌐 v-mem.github.io 🤗 huggingface.co/spaces/liguang… 1/ 🧵

I'm excited for Chuanxia's next step! It was a pleasure to work with him and I couldn't think of a better collaborator and mentor. If you looking for a PhD, don't miss out and apply!

Call for papers: The 26th International Conference on Digital Image Computing: Techniques and Applications (DICTAConference2025) Dates: 3-5 December 2025 Location: Adelaide Convention Centre, Adelaide, Australia Paper Submission: 15 July, 2025 AoE Website: dicta2025.dictaconference.org

Very excited about this! Thank you Runjia for your hard work! If haven’t had the chance, try our demo showcasing our geometry-based memory module for interactive video generators 🕹️🎞️ huggingface.co/spaces/liguang…

Andrea Vedaldi @ ICVSS 2025