Trajectory-Interaction-Graph (TIG)

Auto-generated stub. Edit this file to add more details.

A graph-based module used within a Conditional Flow Matching framework to encode rich spatio-temporal interactions between agents and the scene for trajectory forecasting.

Why It Matters

It explicitly models fine-grained agent-agent and agent-scene relations via a graph structure to enrich the conditional features for the flow-based model.

Evidence

we build a CFM-based predictor with a Trajectory-Interaction-Graph (TIG) module to model fine-grained visual-spatial interactions and strengthen context encoding.

Related Papers

• openalex-2603.24936-tigflow-grpo-trajectory-forecasting-via-interaction-aware-fl