CausalDriveBench A benchmark for causal reasoning in autonomous driving built on top of nuScenes. Each sample bundles a curated causal scene graph , three flavours of multiple choice / open ended QA (active, dormant, distractor), and pointers to the raw nuScenes frames so the benchmark stays compact and license clean. At a glance Subset uploaded: nuscenes Samples: 815 across 475 scenes Tasks: active QA, dormant QA, distractor QA, causal scene graphs Folder structure qa.jsonl is the streaming friendly view (one row = one QA pair with scene id , sample id , qa type , plus the QA fields). The per scene tree under data/ is the authoritative source — use it when you need state / graph / image references for a sample. Resolving image paths Raw nuScenes images are not redistributed — get them from the official nuScenes download page. Once extracted, point NUSCENES ROOT at the dir containing samples/ and sweeps/ : Then, for any sample's frames.json : The data root placeholder lives in one place only : manifest.json nuscenes root placeholder = "${NUSCENES ROOT}" . The per sample frames.json files don't carry it. Loading examples Snapshot the whole repo: Stream only QA via datasets (uses qa.…
We use cookies for essential functionality and analytics. You can accept or reject analytics cookies.Cookie policy