Traj2Video: Code for Video-Based Transportation Mode Recognition from Trajectories

168.9 MB of code for a dual-path framework that converts trajectory data into video representations for machine learning. The resource, authored by Kunkun Fan, was last updated on April 8, 2026. Its core purpose is multiclass transportation mode recognition.

Use Cases

• Train models for transportation mode classification based on trajectory-to-video conversion.
• Benchmark video representation learning techniques for spatiotemporal data.
• Develop frameworks for fusing trajectory and visual data streams for multimodal recognition.

Strengths

• Codebase is 168.9 MB, suggesting substantial implementation details.
• Released under a permissive CC-BY-4.0 license for open reuse and modification.
• Last updated in 2026, indicating recent maintenance.

Limitations

• Description metadata is limited; actual data quality requires manual inspection after download.
• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.

Provenance

Source

Kunkun Fan

Collection Method

Code for a research framework, likely from an academic project.

Time Range

null

Freshness

Last updated 2026-04-08 16:35:38; freshness should be verified.

Geography

null