At #CVPR2023 we're hosting *Scholars and Big Models* -- a forum to discuss recent rapid changes in CV and how our academic community can adapt and thrive. Our panelists will discuss questions raised by you!! Make your voice heard 👇 forms.gle/hVDg4EhrDXuahF…

Seeing the World through Your Eyes paper page: huggingface.co/papers/2306.09… The reflective nature of the human eye is an underappreciated source of information about what the world around us looks like. By imaging the eyes of a moving person, we can collect multiple views of a scene

📢 Learning to Predict Scene-Level Implicit 3D from Posed RGBD Data #CVPR2025. We train a model to predict a 3D implicit function from a single input image. This model is directly trained on raw RGB-D data. Website: nileshkulkarni.github.io/d2drdf Paper: arxiv.org/abs/2306.08671 (1/N)

🚀 We've hit some big milestones with embedchain: • 300K apps • 53K downloads • 5.7K GitHub stars Every step taught us something new. Today, we're taking those lessons & introducing a platform to manage data for LLM apps. No waitlist, link in next tweet 👇🏻👇🏻👇🏻

Annotator disagreement is common in NLP, but is it just noise? We are introducing a new strategy for annotator representation to help models better learn from data that has inherent disagreements. 🐙 Github code: github.com/MichiganNLP/An…

📢 Presenting 𝐅𝐀𝐑: 𝐅𝐥𝐞𝐱𝐢𝐛𝐥𝐞, 𝐀𝐜𝐜𝐮𝐫𝐚𝐭𝐞 𝐚𝐧𝐝 𝐑𝐨𝐛𝐮𝐬𝐭 𝟔𝐃𝐨𝐅 𝐑𝐞𝐥𝐚𝐭𝐢𝐯𝐞 𝐂𝐚𝐦𝐞𝐫𝐚 𝐏𝐨𝐬𝐞 𝐄𝐬𝐭𝐢𝐦𝐚𝐭𝐢𝐨𝐧 #CVPR2024 FAR builds upon complimentary Solver and Learning-Based works yielding accurate *and* robust pose! crockwell.github.io/far/

What do you see in these images? These are called hybrid images, originally proposed by Aude Oliva et al. They change appearance depending on size or viewing distance, and are just one kind of perceptual illusion that our method, Factorized Diffusion, can make.

These spectrograms look like images, but can also be played as a sound! We call these images that sound. How do we make them? Look and listen below to find out, and to see more examples!

In case you were wondering what’s going on with the back of the #CVPR2024 T-shirt: it’s a hybrid image made by Aaron Inbum Park and Daniel Geng! When you look at it up close, you’ll just see the Seattle skyline, but when you view it from a distance, the text “CVPR” should appear.

📢Presenting 𝐃𝐄𝐏𝐈𝐂𝐓: Diffusion-Enabled Permutation Importance for Image Classification Tasks #ECCV2024 We use permutation importance to compute dataset-level explanations for image classifiers using diffusion models (without access to model parameters or training data!)

Self-Supervised Any-Point Tracking by Contrastive Random Walks Ayush Shrivastava, Andrew Owens tl;dr: global matching transformer->self-attention->transition matrix->contrastive random walk->cycle-consistent track arxiv.org/pdf/2409.16288

At #ECCV2024: a very simple, self-supervised tracking method! We train a transformer to perform all-pairs matching using the contrastive random walk. If you want to learn more, please come to our poster at 10:30am on Thursday (#214). w/ Ayush Shrivastava x.com/ayshrv/status/…

We will be presenting this today at #ECCV2024, come say hi. Thursday, 10.30 am - 12.30 pm, poster 214.

Join us at the WorldModelBench workshop at #CVPR2025 where we'll tackle systematic evaluation of World Models! Focus: benchmarks, metrics, downstream tasks, and safety. Submit papers now: worldmodelbench.github.io

Check out Cosmos-Reason1, a reasoning VLM from our team for - Physical Commonsense Reasoning (spatial, temporal, intuitive physics) - Embodied Reasoning (verifying task completion, action affordance and next plausible action prediction) Models, data curation and benchmarks

Video prediction foundation models implicitly learn how objects move in videos. Can we learn how to extract these representations to accurately track objects in videos _without_ any supervision? Yes! 🧵 Work done with: Rahul Venkatesh, Seungwoo (Simon) Kim, Jiajun Wu and Daniel Yamins