Congratulations to our lab colleagues who have been named Outstanding Reviewers at #ICCV2025 👏 Andrei Bursuc @abursuc.bsky.social Anh-Quan Cao @anhquancao.bsky.social Renaud Marlet Eloi Zablocki @eloizablocki.bsky.social @iccv.bsky.social iccv.thecvf.com/Conferences/...

Very happy that finally - after ~2 years of review - PanopticNERF-360 has been accepted to TPAMI :) Panoramic 3D-to-2D Label Transfer in Urban Scenes. Kudos to the team and Yiyi Liao for leading the effort. Here is the project page: fuxiao0719.github.io/projects/pan...

Since we went public a year ago, more than 1.6 mio articles have been read via our paper recommendation platform www.scholar-inbox.com. Super proud of the team! Teaser: Something BIG is coming up soon .. stay tuned!

🚀 Names matter! We show that better class names in open-vocabulary segmentation benchmarks greatly improve dataset quality and boost model performance. RENOVATE your dataset labels with our automatic framework! #AI #ComputerVision #NeurIPS24 andrehuang.github.io/renovate/

Our beginner's oriented accessible introduction to modern deep RL is now published in Foundations and Trends in Optimization. It is a great entry to the field if you want to jumpstart into RL! @bernhard-jaeger.bsky.social www.nowpublishers.com/article/Deta... arxiv.org/abs/2312.08365

Replace "github" with "gitingest" in the url, and you get the whole repo as a single string that you can then paste in your LLMs

Thrilled to announce our workshop on Embodied Intelligence for Autonomous Systems on the Horizon @cvprconference.bsky.social featuring a crazy line-up of speakers and challenges. Mark it in your agendas and also in your registration #cvpr2025 opendrivelab.com/cvpr2025/wor...

The CUDA programming guide docs.nvidia.com/cuda/cuda-c-... is worth studying, even if you're not programming in CUDA. A lot of this knowledge is transferable to DX12/Vulkan compute shaders, especially the GPU architecture/best practices/performance guidelines parts.

Learnings from Scaling Visual Tokenizers for Reconstruction and Generation arxiv.org/abs/2501.09755 The authors study in detail the transition from convolutional backbones to ViT for visual tokenization (ViTok -- excellent name) as well as the impact of scaling up

I wish I stumbled on this pretty and clear intro to CUDA and GPUs years ago : modal.com/gpu-glossary

Excited to share that today our paper recommender platform www.scholar-inbox.com has reached 20k users! We hope to reach 100k by the end of the year.. Lots of new features are being worked on currently and rolled out soon.

That is to say, I think it makes not much sense to fear what others might do. Do the research that you are interested in and I am sure it will be relevant as you bring a unique perspective to the field.

Please tell people that they don’t have to stand out to do good work and be respected. Doing honest research and taking care of your colleagues and your field can be as valuable. We already have so much anxiety and fomo in the field, I’m not sure it’s making us better.

First, new "How to be a good reviewer? Reviewer Tutorial for CVPR 2025" @cvprconference.bsky.social is amazing! docs.google.com/presentation... The thing I don't like - "at least 4 hours if you are not very experienced." I think, 4 - for the "very experienced", and"not exp" should be 6 at least.

The ICCV 2025 website is finally online! 🎉 iccv.thecvf.com @cvprconference.bsky.social @iccv.bsky.social @yoshitomo-cs.bsky.social @csprofkgd.bsky.social @wjscheirer.bsky.social

How do we go beyond colors and recover the intrinsic scene properties? 🤔 👁️ IRIS: Inverse Rendering of Indoor Scenes IRIS estimates accurate material, lighting, and camera response functions given a set of LDR images, enabling photorealistic and view-consistent relighting and object insertion.

I have switched from X and Facebook to Bluesky. I will delete both my X and Facebook accounts in a little bit.

code release for our ECCV paper CarFormer: github.com/Shamdan17/Ca... Shadi is an amazing coder, so this is a great starting point for privileged agents on CARLA. three versions to play with: 1. BEV quantized with a VQ-VAE 2. Attributes (PlanT) 3. BEV => slots (CarFormer)

My book is (at last) out, just in time for Christmas! A blog post to celebrate and present it: francisbach.com/my-book-is-o...

This simple pytorch trick will cut in half your GPU memory use / double your batch size (for real). Instead of adding losses and then computing backward, it's better to compute the backward on each loss (which frees the computational graph). Results will be exactly identical

At some point, we ‘old’ folks who were there before transformers should just gather and write a long survey about everything that was there before…

⚠️Reconstructing sharp 3D meshes from a few unposed images is a hard and ambiguous problem. ☑️With MAtCha, we leverage a pretrained depth model to recover sharp meshes from sparse views including both foreground and background, within mins!🧵 🌐Webpage: anttwo.github.io/matcha/

How to schedule a meeting? When you ask for a meeting with others, you are asking for their time. You are asking for their most valuable, finite resource to benefit yourself (e.g., for advice, networking, questions, and opportunities). Here are some tips that I found useful.

📢 𝐏𝐫𝐄𝐝𝐢𝐭𝐨𝐫𝟑𝐃: 𝐅𝐚𝐬𝐭 𝐚𝐧𝐝 𝐏𝐫𝐞𝐜𝐢𝐬𝐞 𝟑𝐃 𝐒𝐡𝐚𝐩𝐞 𝐄𝐝𝐢𝐭𝐢𝐧𝐠 📢 We propose a training-free 3D shape editing approach that rapidly and precisely edits the regions intended by the user and keeps the rest as is.

📢DNF: Generating 4D animations with dictionary-based neural fields! Xinyi Zhang presents a new dictionary-based neural field for unconditional 4D generation of deforming shapes -- generating motions with high-quality shape and temporal consistency. xzhang-t.github.io/project/DNF/

Here is @anhquancao.bsky.social kicking off his PhD defense on Learning Semantics and Geometry for Scene Understanding in front of a stellar jury. Do check out his great works with quite a bit of impact in this already: anhquancao.github.io/publications/

Proud to announce our NeurIPS spotlight, which was in the works for over a year now :) We dig into why decomposing aleatoric and epistemic uncertainty is hard, and what this means for the future of uncertainty quantification. 📖 arxiv.org/abs/2402.19460 🧵1/10

(1/2) Our proposal 𝐆𝐞𝐧𝟑𝐃: 𝐋𝐞𝐚𝐫𝐧𝐢𝐧𝐠 𝐭𝐨 𝐂𝐫𝐞𝐚𝐭𝐞 𝐕𝐢𝐫𝐭𝐮𝐚𝐥 𝐖𝐨𝐫𝐥𝐝𝐬 was awarded the #ERC Consolidator Grant! The grant comes with 2.75M € in funding and will facilitate cutting-edge research on generative AI methods for 3D content creation.

See who has blocked you 🫣 and added you to a list on Bluesky. clearsky.app

OpenReview feature proposal: Random or chronological order for displaying reviews. Reviews are now displayed in reverse chronological order, meaning that late or emergency reviews are always shown first (potentially getting more attention) at the expense of early reviews that are displayed last.

A librarian that previously worked at the British Library created a relatively small dataset of bsky posts, hundreds of times smaller than previous researchers, to help folks create toxicity filters and stuff. So people bullied him & posted death threats. He took it down. Nice one, folks.

Dear reviewers: As you react/respond to the author rebuttal can you please articulate the answers to these questions in 1-2 sentences each? 1. Why not a lower score 2. Why not a higher score This significantly helps bring everyone (authors/reviewers/AC/SAC) on the same page.

DiffusionDrive: Truncated Diffusion Model for End-to-End Autonomous Driving arxiv.org/abs/2411.151... It learns denoising from anchored Gaussian distribution to multi-node driving action distribution, seemingly without many tricks.

ScaLR is making one step toward 3D foundational models! 🙌 The self-supervised features it learns are quite close to fully-supervised 3D semantic segmentation methods. Still more to do, but quite excited about the progress! 🌟 #CVPR2024

📢 Relive the #CVPR2024 poster ScaLR w/ Gilles Puy. It's a Lidar pretrained model distilled from vision foundation models w/o any labels. 💪 Our best ScaLR is publicly available & well suited for semantic tasks w/ few or no labels & still SoTA results on linear probing www.linkedin.com/posts/andrei...

Every experienced researcher has their own advice for writing scientific papers. Here's mine. users.cecs.anu.edu.au/~sgould/pape...

Check out this BEAUTIFUL interactive blog about cameras and lenses ciechanow.ski/cameras-and-...

Welcome @marcpollefeys.bsky.social ! 🎉

Introducing Generative Omnimatte: A method for decomposing a video into complete layers, including objects and their associated effects (e.g., shadows, reflections). It enables a wide range of cool applications, such as video stylization, compositions, moment retiming, and object removal.

Things you need as a PhD student: - coffee ☕ - a living stipend 💰 - a website 🔗 For the last one, @kiragoldner.bsky.social has good resources: www.kiragoldner.com/blog/website..., www.kiragoldner.com/resources.html

Great article by @melaniemitchell.bsky.social on how confusing it is to talk about LLMs, shown by the wide range of metaphors we draw from Lots of pointers to other work, too www.science.org/doi/full/10....

Charles Darwin's "On the Origin of Species by Means of Natural Selection" was published #OTD in 1859. 🧪 (1/n)

Reviewing is dead. I just read some ICLR reviews. Absolutely junk. No insight. Written by humans, but could have been written by a chatbot or teenager who had read a few dozen reviews. Why do we think this is useful? Only 5% (if that!) of the community has worthwhile opinions.

🚨Hear! Hear! We have a few MSc research internship openings at valeo.ai in Paris for 2025 on computer vision & machine learning (yeah AI). You can find the openings in the link below along with the achievements of our amazing previous interns: valeoai.github.io/interns/ Join us!

ICLR submission (not mine) flagged for Ethics Review based on the use of the term "black box." Reminiscent of the drama involving the term "Byzantine," though there is relevant guidance from ACM this time. openreview.net/forum?id=Z1M...

Self-supervised learning is fantastic for pretraining, but can we use it for other tasks (kNN classification, in-context learning) & modalities, w/o training & by simply using its gradients as features? Enter 🍄FUNGI - Features from UNsupervised GradIents #NeurIPS2024 🧵

NeurIPS Conference is now Live on Bluesky! -NeurIPS2024 Communication Chairs

ICYMI our PointBeV #CVPR2024 poster here's a quick talk by lead author Loïck Chambon. It brings a change of paradigm in multi-camera bird's-eye-view (BeV) segmentation via a flexible mechanism to produce sparse BeV points that can adapt to situation, task, compute www.linkedin.com/posts/andrei...

Discrete diffusion has become a very hot topic again this year. Dozens of interesting ICLR submissions and some exciting attempts at scaling. Here's a bibliography on the topic from the Kuleshov group (my open office neighbors). github.com/kuleshov-gro...

If you have a PhD and you're interested in doing a postdoc near paris on nonsmooth frank Wolfe methods for machine learning, apply for an fmjh post doc here www.fondation-hadamard.fr/fr/programme... deadline is dec 9!