{"success":true,"data":{"author":"Sawtell Lab and Paninski Lab, Columbia University","bucketAccessType":"public","bucketName":"paninski-lab/mirror-mouse-fused","bucketPrefix":"","category":"Neuroscience","cloudProvider":"huggingface","createdAt":"2026-11-21T20:45:58.770Z","description":"A labeled animal pose estimation dataset of head-fixed mice running on a circular treadmill while avoiding a moving obstacle (Warren et al., eLife 2021), recorded by Richard Warren and Qianyun Zhang in the Sawtell Lab at Columbia University. The treadmill has a transparent floor and a 45 degree mirror, so one camera at 250 Hz captures a side view and a bottom view; each frame is labeled with 14 keypoints, seven body points per view with _top and _bot suffixes. This fused single-view version removes the obstacle keypoints of the original. In-distribution and out-of-distribution label sets come from different sessions and animals. Packaged by the Paninski Lab for Lightning Pose: labeled-data/ PNG frames with context frames, CollectedData.csv and CollectedData_test.csv (DeepLabCut-style 2D labels), videos, 51-frame clips around each test frame for the Ensemble Kalman Smoother, a training config and project.yaml. Source archive: Lightning Pose dataset mirror-mouse, DOI 10.6084/m9.figshare.24993315, CC BY 4.0. Cite that record, Warren et al. 2021 and the Lightning Pose paper.","displayTitle":"Mirror Mouse: Lightning Pose Keypoint Dataset of Head-Fixed Mice Running on a Treadmill, Side and Mirror Bottom Views at 250 Hz (Columbia University)","doi":"10.6084/m9.figshare.24993315","huggingFaceUrl":"https://huggingface.co/datasets/paninski-lab/mirror-mouse-fused","id":"32b0bdb7-bb93-5366-971b-4939d288863e","license":"CC-BY-4.0","licenseEvidenceUrl":"https://api.figshare.com/v2/articles/24993315","objectCountEstimate":4902,"probeStatus":"ok","referenceUrl":"https://huggingface.co/datasets/paninski-lab/mirror-mouse-fused","region":"us","schemaCard":{"accessHints":[{"language":"python-datasets","snippet":"from datasets import load_dataset\nds = load_dataset(\"paninski-lab/mirror-mouse-fused\")"},{"language":"python-hub","snippet":"from huggingface_hub import snapshot_download\nsnapshot_download(repo_id=\"paninski-lab/mirror-mouse-fused\", repo_type=\"dataset\")"}],"caveats":["HuggingFace metadata: author=paninski-lab, license:cc-by-4.0, size_categories:n<1K, downloads=599"],"format":"mixed","generatedAt":"2026-11-21T20:48:47.290Z","layout":{"example":"CollectedData.csv","pattern":"{repoId}/{filename}"},"mountPath":"/mnt/archive","sampleHeaders":[{"bytes":"scorer,rick,rick,rick,rick,rick,rick,rick,rick,rick,rick,rick,rick,rick,rick,rick,rick,rick,rick,rick,rick,rick,rick,rick,rick,rick,rick,rick,rick\nbodyparts,paw1LH_top,paw1LH_top,paw2LF_top,paw2LF_top,paw3RF_top,paw3RF_top,paw4RH_top,paw4RH_top,tailBase_top,tailBase_top,tailMid_top,tailMid_top,nose_top,nose_top,paw1LH_bot,paw1LH_bot,paw2LF_bot,paw2LF_bot,paw3RF_bot,paw3RF_bot,paw4RH_bot,paw4RH_bot,tailBase_bot,tailBase_bot,tailMid_bot,tailMid_bot,nose_bot,nose_bot\ncoords,x,y,x,y,x,y,x,y,x,y,x,y,x,y,x,y,x,y,","encoding":"utf8","key":"CollectedData.csv","truncatedAt":512}],"sampleObjects":["CollectedData.csv","CollectedData_test.csv","config_mirror-mouse-fused.yaml","labeled-data/180605_000/img023759.png","labeled-data/180605_000/img023760.png"],"sidecars":[{"bytes":"---\ntask_categories:\n  - keypoint-detection\nlicense: cc-by-4.0\ntags:\n  - biology\n  - pose-estimation\n  - mouse\n  - lightning-pose\npretty_name: Mirror Mouse\nsize_categories:\n  - 1K<n<10K\n---\n\n# Mirror Mouse — Lightning Pose Single-View Dataset\n\nSingle-camera pose estimation dataset for mouse locomotion keypoints, packaged for use with [Lightning Pose](https://github.com/paninski-lab/lightning-pose).\n\n## Dataset Description\n\nHead-fixed mice run on a circular treadmill while avoiding a moving obstacle (Warren et al., *eLife* 2021). The treadmill has a transparent floor and a mirror mounted inside at 45°, allowing a single camera to capture two roughly orthogonal views — a side view and a bottom view via the mirror — simultaneously at **250 Hz**. Each frame is labeled with **14 keypoints**: 7 body keypoints visible from the side (top) and the same 7 keypoints from the bottom (mirror) view.\n\n> **Note:** The original dataset contains keypoints on the moving obstacle; this version removes those keypoints.\n\nSource data: Warren et al., *eLife* 2021 — original archive at https://doi.org/10.6084/m9.figshare.24993315.\nData collected by Richard Warren and Qianyun Zhang in the Sawtell Lab, Columbia University.\n\n## Data Splits\n\n| Split | Labeled frames | Sessions |\n|-------|---------------:|---------:|\n| In-distribution (InD) | 789 | 17 |\n| Out-of-distribution (OOD) | 253 | 5 |\n\nInD and OOD sets contain **different sessions / animals** (no overlap).\n\n- `CollectedData.csv` — InD labels; `videos/` — InD videos\n- `CollectedData_test.csv` — OOD labels; `videos_test/` — OOD videos\n\n## Keypoints\n\n14 keypoints total: 7 body keypoints per view, identified by a `_top` (side view) or `_bot` (mirror/bottom view) suffix.\n\n| Top view (side) | Bottom view (mirror) |\n|-----------------|----------------------|\n| paw1LH_top | paw1LH_bot |\n| paw2LF_top | paw2LF_bot |\n| paw3RF_top | paw3RF_bot |\n| paw4RH_top | paw4RH_bot |\n| tailBase_top | tailBase_bot |\n| tailMid_top | tailMid_bot |\n| nose_top | nose_bot |\n\n## Directory Structure\n\n```\nmirror-mouse-fused/\n├── labeled-data/           # Extracted frames per session; includes ±2 context frames\n├── videos/                 # InD session video clips\n├── videos_test/            # OOD session video clips\n├── videos-for-each-labeled-frame/   # 51-frame videos centered on each OOD labeled frame\n├── CollectedData.csv       # InD 2D keypoint labels (x,y per keypoint)\n├── CollectedData_test.csv  # OOD 2D keypoint labels\n├── config_mirror-mouse-fused.yaml   # Sample Lightning Pose training config\n└── project.yaml            # View and keypoint definitions (required by LP App)\n```\n\nThe `videos-for-each-labeled-frame/` directory contains 51-frame video clips with the labeled frame at the center, intended for use with temporal smoothers such as the [Ensemble Kalman Smoother](https://github.com/paninski-lab/eks).\n\nSee the Lightning Pose documentation for full details on the [single-view data directory structure](https://lightning-pose.readthedocs.io/en/latest/source/directory_structure_reference/singleview_structure.html).\n\n## Usage with Lightning Pose\n\nThe included `config_mirror-mouse-fused.yaml` is a ready-to-use training config. Key settings:\n\n- **Image resize:** 256 × 256\n- **Backbone:** `resnet50_animal_ap10k`\n- **Keypoints:** 14\n- **Mirror columns:** `[0–6]` (top view) and `[7–13]` (bottom view)\n\nUpdate `data.data_dir` to absolute paths on your machine before training.\n\n```bash\nlitpose train config_mirror-mouse-fused.yaml\n```\n\n## Citation\n\nIf you use this dataset, please cite the original paper:\n\n```bibtex\n@article{warren2021rapid,\n  title     = {A rapid whisker-based decision underlying skilled locomotion in mice},\n  author    = {Warren, Richard A and Zhang, Qianyun and Hoffman, Judah R and Li, Edward Y and\n               Hong, Y Kate and Bruno, Randy M and Sawtell, Nathaniel B},\n  journal   = {eLife},\n  volume    = {10},\n  pages     = {e63596},\n  year      = {2021},\n  publisher = {eLife Sciences Publications, Ltd}\n}\n```\n\nOriginal data archive:","encoding":"utf8","key":"README.md","name":"README.md","size":4143,"truncated":true}],"sourceObjectKey":"CollectedData.csv","starterCell":"import pandas as pd\ndf = pd.read_csv(\"hf://datasets/paninski-lab/mirror-mouse-fused/CollectedData.csv\")\ndf.head()","starterSamplePath":"CollectedData.csv","totalObjects":4902},"summary":"A labeled animal pose estimation dataset of head-fixed mice running on a circular treadmill while avoiding a moving obstacle (Warren et al., eLife 2021), recorded by Richard…","title":"Mirror Mouse: Lightning Pose Keypoint Dataset of Head-Fixed Mice Running on a Treadmill, Side and Mirror Bottom Views at 250 Hz (Columbia University)","updatedAt":"2026-11-21T22:07:01.524Z","visibility":"public","websiteUrl":"https://github.com/paninski-lab/lightning-pose"}}