video video 17.2 141 |
|---|
YAML Metadata Warning:The task_categories "imitation-learning" is not in the official list: text-classification, token-classification, table-question-answering, question-answering, zero-shot-classification, translation, summarization, feature-extraction, text-generation, fill-mask, sentence-similarity, text-to-speech, text-to-audio, automatic-speech-recognition, audio-to-audio, audio-classification, audio-text-to-text, voice-activity-detection, depth-estimation, image-classification, object-detection, image-segmentation, text-to-image, image-to-text, image-to-image, image-to-video, unconditional-image-generation, video-classification, reinforcement-learning, robotics, tabular-classification, tabular-regression, tabular-to-text, table-to-text, multiple-choice, text-ranking, text-retrieval, time-series-forecasting, text-to-video, image-text-to-text, image-text-to-image, image-text-to-video, visual-question-answering, document-question-answering, zero-shot-image-classification, graph-ml, mask-generation, zero-shot-object-detection, text-to-3d, image-to-3d, image-feature-extraction, video-text-to-text, keypoint-detection, visual-document-retrieval, any-to-any, video-to-video, other
R2S2R data
This repository preserves the focused data needed to train and evaluate the R2S2R (Scan-to-Sim-to-Real) project:
r2s2r-train300-lerobot.tar.zst: 300 qualified MuJoCo demonstration episodes exported in LeRobot v3 format (100,456 frames, 30 tasks, 20 Hz).r2s2r-frozen-evaluation.tar.zst: frozen Seen-150 and Unseen-150 manifests and their evaluation packages. Evaluation packages intentionally contain no demonstration trajectories.r2s2r-original-scans-v1.tar.zst: seven original Scaniverse GLB inputs and their provenance/hash manifest.SHA256SUMS: archive integrity hashes.
The training observations contain two 256x256 RGB cameras and a seven-element xArm6-plus-gripper state. Actions contain six arm joints plus the gripper. Privileged MuJoCo state, contact, predicate, and object-pose fields were excluded from the policy dataset.
Source project
The code, generation recipes, scene specifications, frozen task definitions, and reproducibility documentation are maintained in the R2S2R source repository. These archives preserve data; they are not a standalone executable environment.
Extraction
sha256sum -c SHA256SUMS
tar --zstd -xf r2s2r-original-scans-v1.tar.zst
tar --zstd -xf r2s2r-train300-lerobot.tar.zst
tar --zstd -xf r2s2r-frozen-evaluation.tar.zst
Licensing and provenance
No single blanket license is asserted for the complete collection. The evaluation packages carry their component license and provenance records, and the scan archive carries the original-scan identity and SHA-256 manifest. Consumers must preserve and follow those embedded terms and verify that their use and redistribution of scan-derived material is authorized.
- Downloads last month
- 43