Yingwei Li

Excited to share two recent works pushing the frontier of 3D perception for physical AI: 🌟 #MPA3D (#CVPR2026 Spotlight) introduces a novel 3D perception model that leverages surfels and 3D Gaussian Splatting (3DGS) to provide rich scene context, removing the dependency on HD maps. Paper: arxiv.org/pdf/2605.22997 📈 #STELLAR is a new #FoundationModel for 3D world perception in physical autonomy. By combining pre-training, mid-training, and fine-tuning, and scaling across data, model size, and tasks, #STELLAR sets a new SOTA on the Waymo Open Dataset benchmark. Paper: arxiv.org/pdf/2605.20390 Huge thanks to @yingwei_li @yangliux1 @yangfu21 @ylzou_Zack and all our incredible co-authors behind these works! Come stop by our poster if you’re attending CVPR next week. #CVPR #AutonomousDriving #PhysicalAI

#iccv We have 3 PhD Research Intern openings to join our Perception Foundation Models team in summer 2026! Interested? Come to ICCV Waymo Booth to talk to intern host @yingwei_li and Song Chen at 2:30PM - 4:30PM, Oct 21 or apply through lnkd.in/gDJQ5ruG

One of the biggest bottlenecks in deploying visual AI and computer vision is annotation, which can be both costly and time-consuming. Today, we’re introducing Verified Auto Labeling, a new approach to AI-assisted annotation that achieves up to 95% of human-level performance while cutting labeling costs by up to 100,000x and time by 5,000x. Read the full paper: arxiv.org/abs/2506.02359

We’re taking the Waymo Driver on another road trip—this time to Washington, D.C., and just in time to catch the cherry blossoms! 🇺🇸🏛️🌸 This testing continues our efforts to introduce our Driver to diverse environments. Learn more about our cross-country training program: waymo.com/blog/2023/11/r…

@CVPR It is worth mentioning that we achieve the SOTA results on Waymo Open Dataset 3D detection leaderboard. By @AdamsYu, @denny_zhou, @quocleix, @YuilleAlan, @tanmingxing et al. @GoogleAI Arxiv:arxiv.org/abs/2203.08195 Leaderboard: bit.ly/36qGtEY↓

Tomorrow (Friday) morning, we will be presenting our @CVPR work 𝐃𝐞𝐞𝐩𝐅𝐮𝐬𝐢𝐨𝐧: 𝐋𝐢𝐝𝐚𝐫-𝐂𝐚𝐦𝐞𝐫𝐚 𝐃𝐞𝐞𝐩 𝐅𝐮𝐬𝐢𝐨𝐧 𝐟𝐨𝐫 𝐌𝐮𝐥𝐭𝐢-𝐌𝐨𝐝𝐚𝐥 𝟑𝐃 𝐎𝐛𝐣𝐞𝐜𝐭 𝐃𝐞𝐭𝐞𝐜𝐭𝐢𝐨𝐧 at 115a. Come visit! Code is just released here shorturl.at/lKU16

📢#CallForPaper AROW Workshop is back @eccvconf! This year we have a great lineup of 9 speakers, and set 3 best paper prizes ($10k each, sponsored by @ftxfuturefund). For more information, please check eccv22-arow.github.io

Learn how DeepFusion, a fully end-to-end multi-modal 3D detection framework, applies a simple yet effective feature fusion strategy to unify the signals from two sensing modalities and achieve state-of-the-art performance. goo.gle/3M2UgRI

DeepFusion: Lidar-Camera Deep Fusion for Multi-Modal 3D Object Detection abs: arxiv.org/abs/2203.08195

LiDAR or camera for self-driving cars? We need both! In our CVPR 2022 paper, we show that multimodality learning can advance state-of-the-art of self-driving car research – our single/ensemble DeepFusion models top the leaderboard of Waymo 3D detection challenge! 1/n

Introducing DeepFusion: a new SoTA lidar-camera multimodal model for autonomous driving. Use cross-attention to align and fuse lidar-camera info, and drastically improve accuracy & robustness for 3D detection. Arxiv:arxiv.org/abs/2203.08195 Leaderboard: bit.ly/36qGtEY↓

We're organizing the workshop on The Art of Robustness: Devil and Angel in Adversarial Machine Learning at #CVPR2022! We have an incredible line-up of speakers and proceedings. The deadline for submission is March 15(AoE). Stay tuned artofrobust.github.io

Happy to announce that we're organizing the workshop on 𝐏𝐫𝐚𝐜𝐭𝐢𝐜𝐚𝐥 𝐃𝐞𝐞𝐩 𝐋𝐞𝐚𝐫𝐧𝐢𝐧𝐠 𝐢𝐧 𝐭𝐡𝐞 𝐖𝐢𝐥𝐝 at #AAAI2022! We have an incredible line-up of speakers and proceedings. The deadline for submission is November 14(AoE). 👉Stay tuned practical-dl.github.io

Our ICCV Workshop on Neural Architectures: Past, Present and Future will start soon (8:55 am EDT, Oct 11), with 7 amazing speakers. Join us via Zoom from @ICCV_2021 platform! We will also be streaming live on Youtube! For more info, please visit neural-architecture-ppf.github.io

𝟐𝐧𝐝 𝐖𝐨𝐫𝐤𝐬𝐡𝐨𝐩 𝐨𝐧 𝐀𝐝𝐯𝐞𝐫𝐬𝐚𝐫𝐢𝐚𝐥 𝐑𝐨𝐛𝐮𝐬𝐭𝐧𝐞𝐬𝐬 𝐢𝐧 𝐭𝐡𝐞 𝐑𝐞𝐚𝐥 𝐖𝐨𝐫𝐥𝐝 @ICCV_2021 happens tomorrow! #ICCV2021 #AROW 🚩Website (join instruction will be here): iccv21-adv-workshop.github.io 📅Time: 08:45 AM – 5:00 PM (EDT, UTC-4), OCTOBER 11, 2021

**Extended deadline** for our 𝟐𝐧𝐝 𝐖𝐨𝐫𝐤𝐬𝐡𝐨𝐩 𝐨𝐧 𝐀𝐝𝐯𝐞𝐫𝐬𝐚𝐫𝐢𝐚𝐥 𝐑𝐨𝐛𝐮𝐬𝐭𝐧𝐞𝐬𝐬 𝐢𝐧 𝐭𝐡𝐞 𝐑𝐞𝐚𝐥 𝐖𝐨𝐫𝐥𝐝 at #ICCV2021! Now due August 5 AoE. iccv21-adv-workshop.github.io

The deadline extends to July 31, 2021, AoE.

We offer several prize and travel grants!! The submission deadline is June 10, 2021, AoE.

Happy to announce that we are organizing the workshop on 𝐀𝐝𝐯𝐞𝐫𝐬𝐚𝐫𝐢𝐚𝐥 𝐋𝐞𝐚𝐫𝐧𝐢𝐧𝐠 𝐟𝐨𝐫 𝐌𝐮𝐥𝐭𝐢𝐦𝐞𝐝𝐢𝐚 at #ACMMM2021! We have an incredible line-up of speakers, and proceedings. Stay tuned! advm-workshop-2021.github.io

Our SecML Workshop at @iclr_conf is tomorrow, with 11 amazing speakers/panelists: @zicokolter, @AdamKortylewski, @aleks_madry, @catherineols, @AlinaMOprea, Aditi Raghunathan, Christopher Ré, @RaquelUrtasun, David Wagner, @YuilleAlan, @ravenben. Link: bit.ly/3b9JQPV