I do have mastodon but it is not as useful as Twitter was once upon a time. I am @violetilla in Blusky, maybe I should give it a proper try. If you want to follow me there. How's the computer vision community over there. Where are people talking nowadays?

This is my Arabic teacher. She's from Gaza, she's real. Suella Braverman has retweeted her to spread hate. Alaa is facing lots of harassment, on top of someone who has been displaced due to genocide and whose family is still trapped in Gaza. Please, share, donate, and help 💜

Statement about our Twitter account and where you can find us instead 👇🏽 See you at @fosdem soon

🔊🔊🔊 TONIGHT

You have until THIS SUNDAY to submit your talk proposals for #FOSDEM2025 We are a friendly & welcoming devroom open to all types of research questions and tooling from an open source/science point of view 🙏🏼 You can submit a 10 or 20min talk ( 5min Q's) research-fosdem.github.io

📢 Excited to share our new work: AC3D: Analyzing and Improving 3D Camera Control in Video Diffusion Transformers snap-research.github.io/ac3d We analyze what pre-trained video diffusion transformers understand about 3D and demonstrate dynamic scene generation with 3D control.

Laura gives another talk in our workshop tomorrow, do not miss it!

🚨 Join Us for the 2nd Workshop on Video Understanding and Its Applications at BMVC 2024! 🗓️ Date: 11/28/2024 🕙 Time: Starts at 10:00 AM GMT 🌍 Web: vua-bmvc.github.io/2024/ Shape the future of video understanding—don’t miss out! @cvssp_research @BMVCconf @PeopleCentredAI

Last week, CoSTAR National Lab was thrilled to welcome local #createch businesses and our partners @SurreyNews to the launch of Futures Studio, our groundbreaking new #virtualproduction facility at @RoyalHolloway. Read more: bit.ly/FuturesStudio #CoSTARNetwork @disguise_one

If you're attending BMVC this year and are passionate about Video Understanding, make sure to join our workshop, VUA, on November 28th. We’re thrilled to feature an incredible lineup of speakers: Prof. @cgmsnoek,Dr @anfurnari, and Dr Laura Sevilla.

This is the behaviour of a rogue state. Germany is a member state of the International Criminal Court. It is legally bound to accept its jurisdiction. Western states are determined to trash what remains of the international order to protect Israel as it commits genocide.

J-PAS has released the most detailed imaging survey of the Universe! It features 54 (!) narrowband filters covering 12 square degrees of the sky (for now, but we will reach 8000!) It is great for #cosmology, #galaxies (both resolved and unresolved), or even #stars! (@aederocl)

👩‍💻🌟 Women in CV 2025: Organizers Application! 🚀Join us in shaping a platform that amplifies voices and fosters connections in computer vision. 💻✨ Fill out the form to apply: forms.gle/8fuarADELxU7XTpW6 #WiCV #CVPR2025 #WomenInSTEM #ComputerVision

The International Criminal Court has issued arrest warrants for Netanyahu & Gallant for war crimes & crimes against humanity. The UK government must comply with the ICC, end all arms sales, impose economic sanctions & suspend its trade deal with Israel. Nothing less will do.

Here we are, and now what?

GPS-Gaussian : Generalizable Pixel-wise 3D Gaussian Splatting for Real-Time Human-Scene Rendering from Sparse Views TL;DR: Are we witnessing the first steps towards 3DGS live streaming? Contributions: • We introduce a generalizable 3D Gaussian Splatting methodology that employs pixel-wise Gaussian parameter maps defined on 2D source image planes to formulate 3D Gaussians in a feed-forward manner. • We propose a fully differentiable framework composed of an iterative depth estimation module and a Gaussian parameter regression module. The intermediate depth prediction bridges the two components and allows them to benefit from joint training. • We introduce a regularization term and an epipolar attention mechanism to preserve geometry consistency between the two source views when using only rendering loss. Our method generalizes well to unseen characters even in complicated scenes. • We develop a real-time FVV system that achieves high-resolution rendering of characters in the scene without any geometry supervision.

Amazing keynote about bringing everyone into the conversation 🙏🏼 exactly what the CoSTAR project aims to do. How do we create AI media production tools *with* artists instead of *for* artists, in a way that is accessible and inclusive for everyone?

Happy to help 🤗

Next up Herbert Potechius from University of Applied Sciences Jena showing us their enhanced illumination adjustment of 3D outdoor reconstructions which improves illuminations changes and removes artefacts using a technique based on shadow removal and colour transfer #CVMP2024

First talk of #CVMP2024 is given by Lewis Bridgeman from @lumirithmic on how to extract high quality facial geometry from few sparse cameras using an active illumination training method that helps with image blur and illumination variance