[CVPR21 WAD] Keynote - Andrej Karpathy, Tesla WAD at CVPR 2021-06-24 | Talk given on 2021/06/20.Andrej is the Senior Director of AI at Tesla, where he leads the team responsible for all neural networks on the Autopilot. Previously, Andrej was a Research Scientist at OpenAI working onDeep Learning in Computer Vision, Generative Modeling and Reinforcement Learning. Andrej received his PhD from Stanford, where he worked with Fei-Fei Li on Convolutional/RecurrentNeural Network architectures and their applications in Computer Vision, Natural Language Processing and their intersection. Over the course of his PhD, Andrej squeezed in two internships at Google where he worked on large-scale feature learning over YouTube videos, and in 2015 he interned at DeepMind and worked on Deep Reinforcement Learning. Together with Fei-Fei, Andrej designed and taught a new Stanford class on Convolutional Neural Networks for Visual Recognition (CS231n). The class was the first Deep Learning course offering at Stanford and has grown from 150 enrolled in 2015 to 330 students in 2016, and 750 students in 2017.
[CVPR26 WAD] Keynote - Vincent Vanhoucke, Waymo WAD at CVPR 2026-09-03 | "Lessons from Driving 200 Million Fully Autonomous Miles"Talk given at the CVPR Workshop on Autonomous Driving 2026: https://cvpr2026.wad.vision/.00:00 Intro03:00 Lesson 1: Scaling as a game of 9s08:47 Lesson 2: Maps are an extremely powerful sensor12:27 Lesson 3: Scaling is the law13:35 Lesson 4: Rare events are merely not frequent yet18:17 Lesson 5: Building a scalable critic is essential20:26 The Waymo Flywheel22:35 Conclusion
[CVPR26 WAD] Keynote - Alexandre Alahi, EPFL WAD at CVPR 2026-09-03 | "3 Principles for Building World Models"Talk given at the CVPR Workshop on Autonomous Driving 2026: https://cvpr2026.wad.vision/.00:00 Intro01:51 Video World Models03:22 Current Challenges04:12 Motion Appearance Decoupling07:52 Infinite-Length Video Generation11:22 LayerSync15:09 SenCache20:09 Limitations and Future Directions
[CVPR26 WAD] Keynote - Holger Caesar, TU Delft WAD at CVPR 2026-09-03 | "Autonomous Driving at the Crossroads"Talk given at the CVPR Workshop on Autonomous Driving 2026: https://cvpr2026.wad.vision/.00:00 Intro01:06 The optimistic view08:24 The pessimistic view23:28 New Datasets and Applications26:40 Conclusion
[CVPR26 WAD] Keynote - Sanja Fidler, NVIDIA WAD at CVPR 2026-09-03 | "Advancing Autonomous Vehicles with World Models"Talk given at the CVPR Workshop on Autonomous Driving 2026: https://cvpr2026.wad.vision/.00:00 Intro01:04 Evolution of Simulation06:28 Simulation 2.0: Neural Reconstruction09:55 Simulation 3.0: Generative World Models11:26 Real-Time World Model (OmniDreams)21:30 OmniDreams Results24:10 OmniDreams as a World-Action Model
[CVPR26 WAD] Keynote - Alex Kendall, Wayve WAD at CVPR 2026-09-03 | "Frontier Challenges in Bringing AI to 1B Robots"Talk given at the CVPR Workshop on Autonomous Driving 2026: https://cvpr2026.wad.vision/.00:00 Intro01:55 Can Autonomy Generalize?05:19 End-to-End AI for Driving09:47 Generative AI for Autonomy12:45 Could we learn to drive in a World Model?14:50 End-to-End should extend to Safety and Simulation too19:07 Platform will reach greater scale than vertically integrated autonomy23:15 Cost matters27:29 Wayve Labs29:24 Conclusion
[CVPR26] Keynote - Shai Shalev-Shwartz, Mobileye WAD at CVPR 2026-09-03 | "Driving the Long Tail: Efficient Scaling via Automatic Scenario Discovery"Talk given at the CVPR Workshop on Autonomous Driving 2026: https://cvpr2026.wad.vision/.00:00 Intro: The Bitter Lesson01:15 Scaling has Diminishing Returns03:38 A Hay in a Haystack07:55 Toward Efficient Scaling: AdaBoost12:30 Scenario Boosting16:15 Scenario Boosting as a Game18:47 Building the Meteor Agent21:05 The Meteor Agent in Action25:51 Conclusion
[CVPR26] Keynote - Ashok Elluswamy, Tesla WAD at CVPR 2026-09-03 | "Building Foundational Models for Robotics at Tesla"Talk given at the CVPR Workshop on Autonomous Driving 2026: https://cvpr2026.wad.vision/.00:00 Intro01:14 Tesla Self-Driving06:35 End-to-end Foundation Model for all of Robotics07:45 The Case for and Challenges in E2E driving11:09 Challenge 1: Curse of Dimensionality15:55 Challenge 2: Interpretability and Safety18:06 Challenge 3: Evaluation21:58 Conclusion
[CVPR26] Keynote - Raquel Urtasun, Waabi WAD at CVPR 2026-09-03 | "Reimagining the Physical World with AI"Talk given at the CVPR Workshop on Autonomous Driving 2026: https://cvpr2026.wad.vision/.00:00 Intro01:10 Design Characteristics of Autonomous Systems02:57 Evolution of Self-Driving07:30 "Next Gen" AV2.011:47 Waabi World16:00 Digital Twinning19:10 The Rise of Generative Models21:18 How can we prove safety?28:06 Conclusion
[CVPR26] Challenge - Argoverse WAD at CVPR 2026-09-03 | Talk given at the CVPR Workshop on Autonomous Driving 2026: https://cvpr2026.wad.vision/.00:00 Intro: What is Argoverse01:45 (Un)supervised Scene Flow07:07 Scene Flow Challenge Winners10:13 Scenario Mining14:20 Scenario Mining Challenge Winners22:00 Future Directions
[CVPR25 WAD] Keynote - Hongyang Li, The University of Hong Kong WAD at CVPR 2025-09-08 | "End-to-end Autonomous Driving: Past, Current and Onwards"Talk given at the CVPR Workshop on Autonomous Driving 2025: https://cvpr2025.wad.vision/.00:00 Introduction00:56 End-to-End Autonomous Driving05:07 Development Stack of Driving Models08:00 World Engine (E2E Driving 2.0)10:28 Nexus: Decoupled Diffusion Sparks Adaptive Scene Generation14:26 MTGS: Multi-Traversal Gaussian Splatting17:07 NAVSim v2: Pseudo-Simulation for Autonomous Driving21:04 Key Challenge: Scalable RL24:18 Takeaways
[CVPR25 WAD] Keynote - Wolfram Burgard, University of Technology Nuremberg WAD at CVPR 2025-09-08 | "Probabilistic and Deep Learning Approaches for Mobile Robotics and Automated Driving"Talk given at the CVPR Workshop on Autonomous Driving 2025: https://cvpr2025.wad.vision/.00:00 Introduction00:48 Trends in AI-based Robotics01:58 Trend 1: Probabilistic Robotics09:53 Trend 2: Deep Networks12:47 Uncertainty-Aware Panoptic Segmentation (EvPSNet)14:54 Uncertainty-Aware LiDAR Panoptic Segmentation (EvLPSNet)16:25 uPLAM: Robust Panoptic Localization & Mapping19:00 Do We Really Need Maps?21:04 Trend 3: Foundation Models22:13 Visual Language Maps for Robot Navigation24:03 FM-Loc: Using Foundation Models for Vision-Based Navigation29:26 Outlook
[CVPR25] Keynote - Laura Leal Taixe, NVIDIA WAD at CVPR 2025-09-08 | "Repurposing Generative Models for 3D Data"Talk given at the CVPR Workshop on Autonomous Driving 2025: https://cvpr2025.wad.vision/.00:00 Introduction01:46 NVIDIA Cosmos Foundation Model04:32 LiDAR Representation05:11 LiDAR Tokenization08:12 Conditions for LiDAR Generation12:31 Joint Modeling of LiDAR & RGB16:25 Final Thoughts
[CVPR25] Keynote - Deva Ramanan, CMU WAD at CVPR 2025-09-08 | "Perception and Simulation for Self-Driving Vehicles"Talk given at the CVPR Workshop on Autonomous Driving 2025: https://cvpr2025.wad.vision/.00:00 Introduction01:12 Section 1: Deep Dive into NuPlan & Planning Metrics04:06 Reactive Agents07:10 State-of-the-Art on NuPlan08:03 Adopting to City-Specific Behavior15:55 Motion Understanding19:03 Utilizing Pseudo-Labels24:02 Scenario Mining26:42 CVPR Scenario Mining Challenge27:29 Conclusion
[CVPR25] Challenge - Argoverse WAD at CVPR 2025-09-08 | Talk given at the CVPR Workshop on Autonomous Driving 2025: https://cvpr2025.wad.vision/.00:00 Introduction00:35 What is Argoverse (2)?02:36 Challenge 1: (Un)supervised Scene Flow05:50 Scene Flow Leaderboard06:23 Winning Method (Delta Flow)08:46 Highlighted Method (Floxels)10:12 Challenge 2: Scenario Mining17:32 Scenario Mining Leaderboard21:03 Challenge 3: Multi-Agent Forecasting22:50 Multi-Agent Forecasting Leaderboard23:56 Future Directions
[CVPR25] Keynote - Wei Zhan, Applied Intuition WAD at CVPR 2025-09-08 | "Scalable Neural Simulation for Autonomy"Talk given at the CVPR Workshop on Autonomous Driving 2025: https://cvpr2025.wad.vision/.00:00 Introduction00:25 Scalable Simulation for E2E Stacks02:28 Gaussian Splatting03:08 DeSiRe-GS: 4D Street Gaussians06:18 Incomplete Objects from 3DGS07:28 R3D2: Realistic 3D Asset Insertion via Diffusion11:36 X-Drive: Cross-Modality Multi-Sensor Data Synthesis15:30 Feed-Forward 3DGS & DrivingRecon18:40 PixelGaussian: Generalizable Gaussian from Arbitrary Views20:00 Onboard and Offboard Models21:26 S2GO: Streaming Sparse Gaussian Occupancy Prediction26:12 Neural Simulator27:25 Conclusion
[CVPR25] Challenge - Nexar WAD at CVPR 2025-09-08 | Talk given at the CVPR Workshop on Autonomous Driving 2025: https://cvpr2025.wad.vision/.00:00 Introduction00:44 About Nexar02:54 Nexar Open Dataset06:02 Nexar Crash Prediction Challenge09:12 Leaderboard11:22 1st Place12:19 2nd Place13:18 3rd Place15:55 Examples of Difficult Cases17:10 Where Next?18:12 Key Takeaways
[CVPR25] Keynote - Chen Wu, Waymo WAD at CVPR 2025-09-08 | "Solving Real-World Challenges of Large-Scale AV Deployment"Talk given at the CVPR Workshop on Autonomous Driving 2025: https://cvpr2025.wad.vision/.00:00 Introduction01:48 Waymo One Service03:30 Challenging Driving Examples04:28 Waymo's Safety Record05:54 The Waymo Foundation Model07:53 Adverse Weather12:47 Sensor Fusion14:22 Occlusion Reasoning19:32 Scene Understanding23:00 Data24:30 Depots - Enabling Operational Excellence25:57 Fast Real-Time Decisions
[CVPR25] Keynote - Xianming Liu, XPeng WAD at CVPR 2025-09-08 | "Scaling up Autonomous Driving via Large Foundation Models"Talk given at the CVPR Workshop on Autonomous Driving 2025: https://cvpr2025.wad.vision/.00:00 Introduction00:40 XPeng's Driving Assistance System02:26 Capabilities in Mass Production03:56 SW 3.0 AI Factory06:45 Foundation Model10:40 Alignment & CoT15:22 Supervised Finetuning (SFT)17:30 Reinforcement Learning19:35 Summary21:23 Additional Demos
[CVPR25] Challenge - Waymo Open Dataset WAD at CVPR 2025-09-08 | Talk given at the CVPR Workshop on Autonomous Driving 2025: https://cvpr2025.wad.vision/.00:00 Introduction01:25 Datasets and Benchmark Overview03:28 2025 Challenges04:43 End-to-End Driving Challenge06:30 New Waymo Dataset of Rare Events09:22 End-to-End Driving Challenge Winners10:08 1st Place: Poutine21:50 Perception Dataset22:41 Motion Dataset23:36 Interaction Prediction Challenge26:19 Interaction Prediction Challenge Winners27:20 1st Place: Parallel ModeSeq32:00 Sim Agents Challenge34:32 Sim Agents Challenge Winners35:05 1st Place: TrajTok40:05 Waymax Update40:30 Scenario Generation Challenge42:07 Scenario Generation Challenge Winners42:50 1st Place: SimFormer49:42 Conclusion
[CVPR25] Workshop on Autonomous Driving (Full Day) WAD at CVPR 2025-09-08 | https://cvpr2025.wad.vision/00:00 Hongyang Li: End-to-end Autonomous Driving: Past, Current and Onwards25:50 Wolfram Burgard: Probabilistic and Deep Learning Approaches for Automated Driving56:00 Laura Leal-Taixe: Repurposing Generative Models for 3D Data01:13:43 Deva Ramanan: Perception and Simulation for Self-Driving Vehicles01:42:26 Argoverse Challenges02:07:13 Wei Zhan: Scalable Neural Simulation for Autonomy02:35:40 Nexar Challenges02:54:52 Chen Wu: Solving Real-World Challenges of Large-Scale AV Deployment03:22:24 Waymo Open Dataset Challenges04:12:53 Xianming Liu: Scaling up Autonomous Driving via Large Foundation Models
[CVPR24 WAD] Keynote - Sanja Fidler, NVIDIA WAD at CVPR 2024-09-12 | Talk given at the CVPR Workshop on Autonomous Driving 2024: https://cvpr2024.wad.vision/.00:00 Introduction01:43 AV Stack Evolution03:43 NVIDIA's Generative AI & Drive Thor05:11 Today's Talk05:41 Foundation Models in Data Tools06:47 Visual Retrieval07:55 Semantic Scenario Clustering08:54 Foundation Model powered AV Stack (AVFM)10:40 PARA-Drive11:10 Hydra-MDP11:41 Trajeglish12:21 Video Generation13:53 Video Modeling16:31 Video Tokenization17:48 Open-Loop Trajectory Prediction18:54 Simulation20:15 Physics, Dynamics and Editing22:00 Closed-Loop Testing23:34 Object Insertion and Handling Large Scenes25:05 Scene Generation25:56 fVDB Release27:28 Conclusion
[CVPR24 WAD] Challenge - Waymo Open Dataset WAD at CVPR 2024-09-12 | Talk given at the CVPR Workshop on Autonomous Driving 2024: https://cvpr2024.wad.vision/.00:00 Introduction02:30 Perception Dataset04:30 Motion Dataset06:22 Benchmarks07:25 Waymax Simulator Release09:18 2024 Challenge Overview10:47 Motion Prediction Challenge15:07 1st Place: MTR v320:12 Occupancy and Flow Challenge23:20 1st Place: DOPP25:11 2nd Place: STNet27:45 3rd Place: HGNET29:57 Sim Agents Challenge33:57 1st Place: BehaviorGPT38:40 Unranked: MPS42:42 3D Semantic Segmentation Challenge45:28 1st Place: Point Transformer V349:51 Conclusion
[CVPR24 WAD] Keynote - Congcong Li, Waymo WAD at CVPR 2024-09-11 | Talk given at the CVPR Workshop on Autonomous Driving 2024: https://cvpr2024.wad.vision/.00:00 Introduction01:52 Waymo's Experience and Service Areas03:26 Driving Examples05:30 Construction, Emergency Vehicles and Freeways09:25 Waymo's Safety Record10:34 Rare Event Examples11:21 Long-Tail Handling12:48 LLM/VLM Reasoning14:13 Open-Vocabulary Perception using VL Distillation19:20 MotionLM: Modeling Driving as a Conversation23:10 MoST: Scene Tokenization for Motion Prediction29:25 Open Questions in the LLM/VLM Space31:25 Waymo Rider Experience32:17 Conclusion
[CVPR24 WAD] Challenge - Argoverse WAD at CVPR 2024-09-11 | Talk given at the CVPR Workshop on Autonomous Driving 2024: https://cvpr2024.wad.vision/.00:00 Introduction00:49 What is Argoverse (2)01:15 Sensor Dataset01:55 LiDAR Dataset02:24 Motion Forecasting Dataset02:55 Map Change Dataset03:57 Argoverse post-Argo05:07 (Un)supervised Scene Flow Challenge18:27 End-to-End Forecasting Challenge32:22 Multi-Agent Motion Forecasting Challenge41:56 4D Occupancy Forecasting Challenge53:32 Conclusion
[CVPR24 WAD] Keynote - Nick Roy, Zoox WAD at CVPR 2024-09-11 | Talk given at the CVPR Workshop on Autonomous Driving 2024: https://cvpr2024.wad.vision/.00:00 Introduction00:59 Autonomy Architectures03:25 Structured Optimization vs E2E Learned Policies06:38 Hybrid Architecture08:34 Early Sensor Fusion11:03 Learned Depth Completion12:45 Handling Debris15:33 The IID Assumption is Brittle16:37 OOD Detection20:06 OOD Detection using Foundation Model Embeddings23:30 Domain Knowledge Helps Create Data25:06 Directed Scenario Generation 28:41 JAX-Accelerated Reinforcement Learning30:30 Symbol Grounding & Conclusion
[CVPR24 WAD] Keynote - Alex Kendall, Wayve WAD at CVPR 2024-09-11 | Talk given at the CVPR Workshop on Autonomous Driving 2024: https://cvpr2024.wad.vision/.00:00 Introduction01:07 The Road to Embodied AI01:57 From Vision to Action03:43 AV2.0: An End-to-End System04:57 Promises of End-to-End Systems06:18 Focus Areas to Create Embodied AI06:53 Focus Area: Simulation08:36 Ghost Gym: Neural Simulator09:28 Introducing PRISM-112:09 Launching WayveScenes10114:07 GAIA: Generative World Model18:13 Focus Area: Multimodality19:27 LINGO-1: Offline Video QA21:21 LINGO-2: Grounded Vision-Language-Action Model24:32 AV2.0 Reimagining the Driving Experience25:30 Focus Area: Engineering to Product28:00 Embodied AI Foundation Models31:38 Safety by Design33:05 Conclusion
[CVPR24 WAD] Keynote - Bolei Zhou, UCLA WAD at CVPR 2024-09-11 | Talk given at the CVPR Workshop on Autonomous Driving 2024: https://cvpr2024.wad.vision/.00:00 Introduction00:52 The Gap between Industry and Academia02:02 Driving Datasets02:51 Driving Simulators04:41 The MetaDrive Simulator07:26 ScenarioNet11:29 TrafficGen13:30 Closed-Loop Adversarial Training (CAT)16:18 SimGen Driving Scene Generation21:04 Mobility Anywhere22:55 MetaUrban Simulator25:35 Conclusion
[CVPR24 WAD] Keynote - Kashyap Chitta, University of Tübingen WAD at CVPR 2024-09-11 | Talk given at the CVPR Workshop on Autonomous Driving 2024: https://cvpr2024.wad.vision/.00:00 Introduction00:30 Benchmarking AVs is hard02:36 Flaws in Open Loop Benchmarks05:18 What about Simulation07:18 Non-reactive Simulation08:11 No Sensor Simulation11:54 No Traffic Simulation13:33 NAVSIM Metrics16:00 The Predictive Driver Model (PDM) Score18:10 Comparing PDMS vs Closed-loop Metrics20:09 OpenScene Release to make E2E Research more Accessible24:15 OpenScene Benchmark Results25:54 2024 NAVSIM Challenge26:35 Limitations27:56 Conclusion
[CVPR24 WAD] Keynote - Georgios Pavlakos WAD at CVPR 2024-09-11 | Talk given at the CVPR Workshop on Autonomous Driving 2024: https://cvpr2024.wad.vision/.00:00 Introduction00:58 Perceiving Humans in 4D03:17 Improving Robustness in Human Motion Estimation04:49 Human Mesh Recovery (HMR)07:46 HMR Examples11:28 Application for Tracking13:47 SLAHMR: Decoupling Human and Camera Motion in the Wild18:36 SLAHMR Examples21:30 Reconstructing Humans and Scenes22:24 Reconstructing 3D Humans in Contact22:55 Reasoning about Humans in Simulation23:31 Building Realistic Avatars24:00 Conclusion
[CVPR23 WAD] Keynote - Hang Zhao, Tsinghua University WAD at CVPR 2023-06-30 | Talk given at the CVPR Workshop on Autonomous Driving 2023: https://cvpr2023.wad.vision/.00:00 Introduction00:27 A Simplified Self-Driving Stack01:05 ViP3D: End-to-End Visual Prediction02:25 Scalability04:56 3D Occupancy Prediction07:28 Auto-Labeling Occupancy Datasets12:25 The Occ3D and SSCBench Benchmarks13:47 Handling New Geo-Locations16:31 VectorMapNet18:59 Neural Map Priors22:10 Map Prior Improving Range and Robustness23:35 Conclusion
[CVPR23 WAD] Challenge - Argoverse WAD at CVPR 2023-06-29 | Talk given at the CVPR Workshop on Autonomous Driving 2023: https://cvpr2023.wad.vision/.00:00 Introduction01:50 What is Argoverse (2)7:17 Multi-Agent Motion Forecasting Challenge20:57 End-to-End Forecasting Challenge36:26 4D Occupancy Forecasting Challenge44:38 Self-Supervised Scene Flow Challenge
[CVPR23 WAD] Challenge - BDD100K WAD at CVPR 2023-06-29 | Talk given at the CVPR Workshop on Autonomous Driving 2023: https://cvpr2023.wad.vision/.00:00 Introduction00:35 First Winner12:35 Second Winner
[CVPR23 WAD] Keynote - Andreas Wendel, Kodiak Robotics WAD at CVPR 2023-06-29 | Talk given at the CVPR Workshop on Autonomous Driving 2023: https://cvpr2023.wad.vision/.
[CVPR23 WAD] Challenge - Waymo Open Dataset WAD at CVPR 2023-06-29 | Talk given at the CVPR Workshop on Autonomous Driving 2023: https://cvpr2023.wad.vision/.00:00 Introduction06:15 What's New08:04 Waymax & Perception Object Assets11:13 New Dataset Format14:43 Polling21:47 Pose Estimation Challenge31:57 Motion Prediction Challenge42:30 Sim Agents Challenge54:42 2D Video Panoptic Segmentation
[CVPR23 WAD] Keynote - Alexandre Alahi, EPFL WAD at CVPR 2023-06-29 | Talk given at the CVPR Workshop on Autonomous Driving 2023: https://cvpr2023.wad.vision/.00:00 Introducing Social Forecasting03:05 Robot Experiment04:34 Representation Learning: Perception08:52 Representation Learning: Social Forecasting16:57 Representation Learning: Planning19:13 7 Foundational Principles for Autonomous Mobility19:22 P1: Predictive Coding21:18 P2: Opposites23:30 P3: Multimodality24:09 P4: Focused Learning25:21 P5: Compositionality and Sparsity26:41 P6: Causality28:07 P7: In Vivo Practice29:11 Summary
[CVPR23 WAD] Keynote - Ashok Elluswamy, Tesla WAD at CVPR 2023-06-29 | Talk given at the CVPR Workshop on Autonomous Driving 2023: https://cvpr2023.wad.vision/.00:00 Introduction02:09 Occupancy Networks Recap04:04 Generative Modeling of Lanes06:15 Object Prediction & Properties08:10 Fleet Auto-Labeling13:23 Learning a General World Model18:04 Large Compute20:02 Q&A
[CVPR23 WAD] Keynote - Jiyang Gao, Momenta WAD at CVPR 2023-06-29 | Talk given at the CVPR Workshop on Autonomous Driving 2023: https://cvpr2023.wad.vision/.00:00 Introduction00:55 Long-Tail Problem02:11 Momenta Flywheel Strategy02:57 Momenta's Mapless Solution04:55 DDLD: Data Driven Landmark Detection10:48 DLP: Deep Learning Planning17:35 DDPF: Data Driven Pose Fusion25:25 Summary
[CVPR23 WAD] Keynote - Chelsea Finn, Stanford University/Google WAD at CVPR 2023-06-29 | Talk given at the CVPR Workshop on Autonomous Driving 2023: https://cvpr2023.wad.vision/.00:00 Introduction01:25 Examples of Distribution Shift03:48 Takeaways from Benchmark Creation Process08:05 What Should We Do About Distribution Shift?08:49 Detecting Distribution Shift19:51 Improve Predictions under Shift24:06 Diversify and Disambiguate (DivDis)26:50 Overall Takeaways
[CVPR23 WAD] Keynote - Andreas Geiger, Universität Tübingen WAD at CVPR 2023-06-29 | Talk given at the CVPR Workshop on Autonomous Driving 2023: https://cvpr2023.wad.vision/.00:00 Introduction03:12 nuPlan04:53 Open-Loop and Closed-Loop05:57 Lesson 1: A route centerline is all you need for ego-forecasting08:25 Lesson 2: Rule-based beat learned planners in closed-loop10:06 Lesson 3: Planning & ego-forecasting tasks are misaligned16:01 Lesson 4: Learned/long-term forecasting doesn't improve closed-loop17:49 CARLA22:00 TransFuser22:58 Lesson 5: Target point conditioned methods learn shortcut when out-of-distribution25:26 Lesson 6: Average pooling remove spatial info and increases bias27:11 Lesson 7: Waypoint representations work well as they learn to interpolate28:48 Summary
[CVPR23 WAD] Keynote - Chen Wu, Waymo WAD at CVPR 2023-06-29 | Talk given at the CVPR Workshop on Autonomous Driving 2023: https://cvpr2023.wad.vision/.0:00 Introduction0:21 Waymo's Progress3:36 Adverse Weather5:21 Navigating Construction Zones6:49 Handling Dense Traffic7:35 Safety10:28 Core Architecture: MotionDiffuser14:11 Core Architecture: SWFormer17:37 Long-Tail: Rare Example Mining21:04 Long-Tail: Open-Set Perception24:56 Holistic Approach to Handling Adverse Weather
[CVPR22 WAD] Keynote - Alex Kendall, Wayve WAD at CVPR 2022-08-20 | Talk given at the Workshop on Autonomous Driving at CVPR 2022.2022-06-20
[CVPR22 WAD] Keynote - Sergey Levine, UC Berkeley WAD at CVPR 2022-08-20 | Talk given at the Workshop on Autonomous Driving at CVPR 2022.2022-06-20
[CVPR22 WAD] Keynote - Yuning Chai, Cruise WAD at CVPR 2022-08-20 | Talk given at the Workshop on Autonomous Driving at CVPR 2022.2022-06-20
[CVPR22 WAD] Challenge - Argoverse WAD at CVPR 2022-08-20 | Presented at the Workshop on Autonomous Driving at CVPR 2022.2022-06-20
[CVPR22 WAD] Challenge - nuPlan by Motional WAD at CVPR 2022-08-20 | Presented at the Workshop on Autonomous Driving at CVPR 2022.2022-06-20
[CVPR22 WAD] Challenge - BDD100K WAD at CVPR 2022-08-20 | Presented at the Workshop on Autonomous Driving at CVPR 2022.2022-06-20
[CVPR22 WAD] Challenge - Waymo Open Dataset WAD at CVPR 2022-08-20 | Presented at the Workshop on Autonomous Driving at CVPR 2022.2022-06-20
[CVPR22 WAD] Keynote - Matthew Johnson-Roberson, CMU WAD at CVPR 2022-08-20 | Talk given at the Workshop on Autonomous Driving at CVPR 2022.2022-06-20
[CVPR22 WAD] Keynote - Raquel Urtasun, Waabi/UofT WAD at CVPR 2022-08-20 | Talk given at the Workshop on Autonomous Driving at CVPR 2022.2022-06-20
[CVPR22 WAD] Keynote - Sanja Fidler, Nvidia/UofT WAD at CVPR 2022-08-20 | Talk given at the Workshop on Autonomous Driving at CVPR 2022.2022-06-20
![[CVPR21 WAD] Keynote - Andrej Karpathy, Tesla](https://i.ytimg.com/vi/g6bOwQdCJrc/hqdefault.jpg)