-
X-Sense:
Ego-Exo Sensing for Smart Mobility Workshop
In Conjunction with the International Conference on Computer Vision 2025
Hawai'i Convention Center @ Room 323C
October 20, 2025
Overview
This workshop explores a holistic approach to ego-exo sensing, integrating vehicle sensors, roadside cameras, aerial imagery, and V2V communications to advance transportation intelligence and drive progress toward smart mobility. We examine how ego-exo sensing networks enhance safety-critical scenario detection and generation, comprehensive environmental perception, cooperative driving, and multi-agent decision making, among other crucial tasks shaping the future of mobility. This workshop bridges siloed research efforts to create unified approaches for heterogeneous sensor fusion that will define next-generation mobility systems.
Keynote Speakers
Schedules
| Start Time | End Time | Event |
|---|---|---|
| 09:00 | 09:10 | Opening remarks |
| 09:10 | 09:40 | Invited talk #1 – Prof. Bharath Hariharan Topic: Seeing through The Eye in the Sky |
| 09:40 | 10:10 | Invited talk #2 – Prof. Cathy Wu Topic: Mitigating Metropolitan Carbon Emissions with Dynamic Eco-driving at Scale |
| 10:10 | 10:30 | Oral Presentation #1 1. ArgoTweak: Towards Self-Updating HD Maps through Structured Priors 2. QuantV2X: A Fully Quantized Multi-Agent System for Cooperative Perception |
| 10:30 | 10:50 | Coffee break |
| 10:50 | 11:20 | Invited talk #3 – Prof. Manmohan Chandraker Topic: Towards Agentic Autonomy |
| 11:20 | 11:40 | Oral Presentation #2 1. TurboTrain: Towards Efficient and Balanced Multi-Task Learning for Multi-Agent Perception and Prediction 2. InSPE: Rapid Evaluation of Heterogeneous Multi-Modal Infrastructure Sensor Placement |
| 11:40 | 12:30 | Poster Section |
| 12:30 | 13:30 | Lunch break |
| 13:30 | 14:00 | Invited talk #4 – Dr. Katie Z Luo Topic: Leveraging Auxiliary Information for Self-Driving Perception |
| 14:00 | 14:30 | Invited talk #5 – Prof. Zhengzhong Tu Topic: Agentic V2X Systems: Towards Smarter, Safer, and More Trustworthy Connected Mobility |
| 14:30 | 15:00 | Invited talk #6 – Prof. Bolei Zhou Topic: Scaling Physical AI for Sidewalk Autonomy |
| 15:00 | 15:30 | Coffee break |
| 15:30 | 16:00 | Invited talk #7 – Dr. Boyi Li Topic: Learning to simulate multimodal robot worlds |
| 16:00 | 17:00 | Panel on future ego-exo sensing for smart mobility |
| 18:00 | 20:00 | Social Networking and Dinner (Location will be announced during the workshop) |
Accepted Papers
Oral Papers:
- ArgoTweak: Towards Self-Updating HD Maps through Structured Priors
- TurboTrain: Towards Efficient and Balanced Multi-Task Learning for Multi-Agent Perception and Prediction
- QuantV2X: A Fully Quantized Multi-Agent System for Cooperative Perception
- InSPE: Rapid Evaluation of Heterogeneous Multi-Modal Infrastructure Sensor Placement
Poster Papers:
- O-MaMa: Learning Object Mask Matching between Egocentric and Exocentric Views
- Learning 3D Perception from Others' Predictions
- Enhancing Text-to-Image Retrieval with VLM Reranking via Contrastive Reinforcement Learning
- MamV2XCalib: V2X-based Target-less Infrastructure Camera Calibration with State Space Model
- PhraseStereo: The First Open-Vocabulary Stereo Image Segmentation Dataset
- EgoExoBench: A Benchmark for First- and Third-person View Video Understanding in MLLMs
- RelMap: Enhancing Online Map Construction with Class-Aware Spatial Relation and Semantic Priors
- PLEUM: A Framework for Benchmarking Ego-Exo Perception Gaps in Polish Urban Mobility
- SmartCityZero: Label-Free 3D Perception through RSU-to-Vehicle Collaboration
- SAGE: A Synchronized Action and Gaze Estimation Framework for Ego-Exo Human Behavior Analysis
- Ego-ExoFusion: Benchmarking Multi-Agent Interaction and Trajectory Prediction using Integrated Egocentric and Exocentric Views
- Beyond Pixels: Metadata Fusion for Enhanced Driving Scene Segmentation
- AirV2X: Unified Air-Ground Vehicle-to-Everything Collaboration
- V2V-GoT: Vehicle-to-Vehicle Cooperative Autonomous Driving with Multimodal Large Language Models and Graph-of-Thoughts
- RadarSplat: Radar Gaussian Splatting for High-Fidelity Data Synthesis and 3D Reconstruction of Autonomous Driving Scenes
Call for Papers
We invite high-quality submissions addressing integrated ego-exo sensing for smart mobility. Relevant topics include, but are not limited to:
- Collection of synchronized egocentric and exocentric data for smart mobility
- Detection and generation of safety-critical and edge-case scenarios via ego-exo sensing
- Efficient annotation of egocentric and exocentric data
- Privacy-preserving and anonymization techniques for cross-sensor data
- Multimodal large models for scene understanding, cooperative driving, and multi-agent decision-making
- Modeling and simulation of heterogeneous traffic behavior via integrated ego-exo sensing modalities
- Data generation for training and testing cooperative driving systems
- Modeling of multi-agent interaction and decision processes
- Benchmarking safety, pollution, and efficiency in smart mobility systems
Submission Instructions
We offer a flexible platform for presenting a broad range of contributions:
- Extended Abstracts and Short Papers: Submissions of up to 4 pages (excluding references) are welcome, including work in progress, negative results with value to the community, or thought-provoking position papers on relevant topics.
- Previously Published Work: We also invite submissions of papers previously published elsewhere, including those accepted to the main ICCV 2025 conference. Submissions of up to 8 pages (excluding references) are welcome.
All submissions should follow the ICCV 2025 formatting guidelines . Accepted papers will be presented at the workshop as oral presentations or posters.
Important Dates
- Submission Deadline: September 8
August 31, 2025 09:59 AM GMT - Preliminary Author Notification: September 21
14, 2025 - Submission Portal: OpenReview
