Session Abstract

This special session focuses on event driven and multimodal perception in low altitude, land and waterborne transportation. Modern transportation environments contain fast motion, changing illumination, adverse weather, degraded image quality and strict latency and energy requirements. These challenges make robust multimedia perception a key requirement for safe and reliable systems. New sensing modalities such as event cameras, spike cameras, infrared imaging, polarization imaging, hyperspectral imaging and radar vision fusion offer higher temporal resolution and better environmental adaptability. At the same time, advances in multimedia understanding including representation learning, video analysis, generative modeling and vision language models are improving perception under complex and low quality conditions. The session aims to bring together research on visual and multimodal perception for all transportation scenarios and welcomes work on any single modality or multimodal fusion that improves multimedia understanding in realistic and challenging environments.

Call for Papers

The Special Session on Event Driven and Multimodal Perception in Low Altitude, Waterborne and Ground Transportation invites original research papers that advance multimedia sensing and understanding in real world transportation environments. Submissions should focus on visual or multimodal perception for aerial, land based or maritime scenarios, including both single modality methods and multimodal fusion. Accepted papers will be included in ICME 2026 and presented in the special session. Researchers from multimedia, computer vision and intelligent transportation communities are encouraged to submit.

Topics of interest include (but are not limited to):

• Multimedia perception for low-altitude, land, and waterborne transportation
• RGB-based, neuromorphic-based, IR/thermal, polarization, or hyperspectral visual analysis
• New imaging modalities and multimedia sensing technologies for transportation
• Visual scene reconstruction and environmental modeling in low-altitude, land, and waterborne environments using traditional or emerging visual sensing modalities
• Robust multimedia understanding under degradation, noise, or adverse conditions
• Representation learning, domain generalization, and cross-view or cross-modal alignment
• Video content understanding and high-speed perception for safety-critical tasks
• Multimedia-based behavior, intention, or anomaly understanding
• Multimedia datasets, benchmarks, and real-world deployments
• Generative models for enhancement, reconstruction, simulation, and content synthesis
• Foundation models and vision-language models for transportation multimedia
• Multimedia perception for autonomous and embodied agents across air, land, and water
• Safety, robustness, privacy, and security in multimedia-driven transportation systems
• Applications in low-altitude logistics, smart highways, ports, inland waterways, and cross-domain traffic

programme

TBD

• Papers must be no longer than 6 pages, including all text, figures, and references.
• The abstract should be about 100-150 words.
• You can obtain more detailed information through this link.

Important Reminder: When submitting via CMT, please select our Special Session as the "Primary Subject Area" (or Track) to ensure your paper is routed to our review.