l5kit
l5kit copied to clipboard
Some questions about AgentDataset and visualization
Hi, dear L5kit team,
May I ask the question that why 'select_agent.py’ use the 'th_distance_av' threshold to mask some of the observed agents? In my understanding, the quality of other agent states has been guaranteed by other thresholds.
Another question, is there any existing API that can be used to visualized a scenario? Existing API seems likely to visualize the context around an agent, instead of a whole scene.
thank you.