Add files using upload-large-folder tool
Browse files- Surgical/utenn/benchtop_datasets_round2_with_part_seg/README.md +227 -0
- Surgical/utenn/benchtop_datasets_round2_with_part_seg/meta/episodes.jsonl +2 -0
- Surgical/utenn/benchtop_datasets_round2_with_part_seg/meta/episodes_stats.jsonl +2 -0
- Surgical/utenn/benchtop_datasets_round2_with_part_seg/meta/info.json +75 -0
- Surgical/utenn/benchtop_datasets_round2_with_part_seg/meta/tasks.jsonl +1 -0
- Surgical/utenn/surgical_video_datasets_round2_with_part_seg/videos/chunk-000/observation.images.tool_segmentation_id/episode_000005.mkv +0 -0
- Surgical/utenn/surgical_video_datasets_round2_with_part_seg/videos/chunk-000/observation.images.tool_segmentation_id/episode_000015.mkv +0 -0
- Surgical/utenn/surgical_video_datasets_round2_with_part_seg/videos/chunk-000/observation.images.tool_segmentation_id/episode_000029.mkv +0 -0
- Surgical/utenn/surgical_video_datasets_round2_with_part_seg/videos/chunk-000/observation.images.tool_segmentation_id/episode_000039.mkv +0 -0
- Surgical/utenn/surgical_video_datasets_round2_with_part_seg/videos/chunk-000/observation.images.tool_segmentation_id/episode_000056.mkv +0 -0
- Surgical/utenn/surgical_video_datasets_round2_with_part_seg/videos/chunk-000/observation.images.tool_segmentation_id/episode_000073.mkv +0 -0
- Surgical/utenn/surgical_video_datasets_round2_with_part_seg/videos/chunk-000/observation.images.tool_segmentation_id/episode_000077.mkv +0 -0
- Surgical/utenn/surgical_video_datasets_round2_with_part_seg/videos/chunk-000/observation.images.tool_segmentation_id/episode_000078.mkv +0 -0
- Surgical/utenn/surgical_video_datasets_round2_with_part_seg/videos/chunk-000/observation.images.tool_segmentation_id/episode_000080.mkv +0 -0
- Surgical/utenn/surgical_video_datasets_round2_with_part_seg/videos/chunk-000/observation.images.tool_segmentation_id/episode_000083.mkv +0 -0
- Surgical/utenn/surgical_video_datasets_round2_with_part_seg/videos/chunk-000/observation.images.tool_segmentation_id/episode_000084.mkv +0 -0
- Surgical/utenn/surgical_video_datasets_round2_with_part_seg/videos/chunk-000/observation.images.tool_segmentation_id/episode_000093.mkv +0 -0
- Surgical/utenn/surgical_video_datasets_round2_with_part_seg/videos/chunk-000/observation.images.tool_segmentation_id/episode_000094.mkv +0 -0
- Surgical/utenn/surgical_video_datasets_round2_with_part_seg/videos/chunk-000/observation.images.tool_segmentation_id/episode_000098.mkv +0 -0
- Surgical/utenn/surgical_video_datasets_round2_with_part_seg/videos/chunk-000/observation.images.tool_segmentation_id/episode_000100.mkv +0 -0
Surgical/utenn/benchtop_datasets_round2_with_part_seg/README.md
ADDED
|
@@ -0,0 +1,227 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
<!--
|
| 2 |
+
Open-H Embodiment Dataset README Template (v1.0)
|
| 3 |
+
Please fill out this template and include it in the ./metadata directory of your LeRobot dataset.
|
| 4 |
+
This file helps others understand the context and details of your contribution.
|
| 5 |
+
-->
|
| 6 |
+
|
| 7 |
+
# Benchtop Surgical Tool Manipulation Dataset (Open-H) - README
|
| 8 |
+
|
| 9 |
+
---
|
| 10 |
+
|
| 11 |
+
## 📋 At a Glance
|
| 12 |
+
|
| 13 |
+
Bench-top surgical-tool video episodes with aligned **RGB**, **per-frame depth** (NPZ + visualization video), **part segmentation label maps**, **2D tool poses**, and optional **dynamic 3D Gaussian** scene/tool representations.
|
| 14 |
+
|
| 15 |
+
---
|
| 16 |
+
|
| 17 |
+
## 📖 Dataset Overview
|
| 18 |
+
|
| 19 |
+
This dataset packages labeled benchtop clips into a **LeRobot dataset v2.1-style** directory layout that is **submit-ready for Open-H**. Each episode corresponds to a single recorded clip under an action label (task). For every timestep, we provide consistent indexing and references to raw depth/segmentation assets, plus per-tool 2D pose states.
|
| 20 |
+
|
| 21 |
+
| | |
|
| 22 |
+
| :--- | :--- |
|
| 23 |
+
| **Total Trajectories** | `2` |
|
| 24 |
+
| **Total Hours** | `Around 25 seconds` |
|
| 25 |
+
| **Data Type** | `[ ] Clinical` `[ ] Ex-Vivo` `[x] Table-Top Phantom` `[ ] Digital Simulation` `[ ] Physical Simulation` `[ ] Other (If checked, update "Other")` |
|
| 26 |
+
| **License** | CC BY 4.0 |
|
| 27 |
+
| **Version** | `1.0` |
|
| 28 |
+
|
| 29 |
+
---
|
| 30 |
+
|
| 31 |
+
## 🧭 Dataset Structure & Navigation
|
| 32 |
+
|
| 33 |
+
This dataset follows the LeRobot v2.1-style layout (manual writer) with **videos**, **per-episode parquet tables**, and **raw assets**.
|
| 34 |
+
|
| 35 |
+
```
|
| 36 |
+
<dataset_root>/
|
| 37 |
+
data/
|
| 38 |
+
chunk-000/
|
| 39 |
+
episode_000000.parquet
|
| 40 |
+
episode_000001.parquet
|
| 41 |
+
...
|
| 42 |
+
videos/
|
| 43 |
+
chunk-000/
|
| 44 |
+
observation.images.rgb/ episode_000000.mp4 ...
|
| 45 |
+
observation.images.depth_vis/ episode_000000.mp4 ...
|
| 46 |
+
observation.images.part_id/ episode_000000.mp4 ...
|
| 47 |
+
meta/
|
| 48 |
+
info.json
|
| 49 |
+
tasks.jsonl
|
| 50 |
+
episodes.jsonl
|
| 51 |
+
episodes_stats.jsonl
|
| 52 |
+
metadata/
|
| 53 |
+
README.md <-- this file
|
| 54 |
+
raw/
|
| 55 |
+
depth/episode_000000/frame_000000.npz ...
|
| 56 |
+
seg/episode_000000/00000.png ...
|
| 57 |
+
toolposes/episode_000000/tool_poses.json
|
| 58 |
+
toolposes/episode_000000/tool_name_to_slot.json
|
| 59 |
+
seg_schema/episode_000000/label_schema.json
|
| 60 |
+
seg_schema/episode_000000/label_id_maps.json
|
| 61 |
+
gaussians/episode_000000/dynamic_gaussians_frames.jsonl.gz (optional)
|
| 62 |
+
gaussians/episode_000000/dynamic_gaussians_meta.json (optional)
|
| 63 |
+
gaussians/episode_000000/photometric_cam_T_wc.json (optional)
|
| 64 |
+
```
|
| 65 |
+
|
| 66 |
+
### Where to look first
|
| 67 |
+
- **meta/info.json**: overall dataset metadata (fps, splits, modalities, counts)
|
| 68 |
+
- **meta/tasks.jsonl**: list of action labels / tasks
|
| 69 |
+
- **meta/episodes.jsonl**: episode index → clip id, paths, tool list, gaussians presence
|
| 70 |
+
- **data/chunk-000/episode_XXXXXX.parquet**: per-timestep table (timestamps, tool poses, raw pointers)
|
| 71 |
+
- **videos/**: fast visualization for RGB / depth_vis / part_id
|
| 72 |
+
- **raw/**: submission-grade assets (relative depth NPZ, label-map PNGs, toolposes JSON, optional gaussians)
|
| 73 |
+
|
| 74 |
+
---
|
| 75 |
+
|
| 76 |
+
## 🎯 Tasks & Domain
|
| 77 |
+
|
| 78 |
+
### Domain
|
| 79 |
+
- [x] **Surgical Robotics**
|
| 80 |
+
- [ ] **Ultrasound Robotics**
|
| 81 |
+
- [ ] **Other Healthcare Robotics** (Please specify: `N/A`)
|
| 82 |
+
|
| 83 |
+
### Demonstrated Skills
|
| 84 |
+
Action labels are stored in:
|
| 85 |
+
- `meta/tasks.jsonl` (task definitions)
|
| 86 |
+
- `action.label` column in each parquet
|
| 87 |
+
|
| 88 |
+
Examples (replace with your actual labels):
|
| 89 |
+
- Tool manipulation (grasp / pick-and-place)
|
| 90 |
+
- Retraction / interaction
|
| 91 |
+
- Cutting / dissection-like motions
|
| 92 |
+
|
| 93 |
+
---
|
| 94 |
+
|
| 95 |
+
## 🔬 Data Collection Details
|
| 96 |
+
|
| 97 |
+
### Collection Method
|
| 98 |
+
- [x] **Human Teleoperation** *(typical for benchtop surgical tool demos; update if different)*
|
| 99 |
+
- [ ] **Programmatic/State-Machine**
|
| 100 |
+
- [ ] **AI Policy / Autonomous**
|
| 101 |
+
- [ ] **Other** (Please specify: `TBD`)
|
| 102 |
+
|
| 103 |
+
### Operator Details
|
| 104 |
+
|
| 105 |
+
| | Description |
|
| 106 |
+
| :--- | :--- |
|
| 107 |
+
| **Operator Count** | `N/A` |
|
| 108 |
+
| **Operator Skill Level** | `[ ] Expert (e.g., Surgeon, Sonographer)` <br> `[x] Intermediate (e.g., Trained Researcher)` <br> `[ ] Novice (e.g., ML Researcher with minimal experience)` <br> `[ ] N/A` |
|
| 109 |
+
| **Collection Period** | `N/A` |
|
| 110 |
+
|
| 111 |
+
### Recovery Demonstrations
|
| 112 |
+
- [ ] **Yes**
|
| 113 |
+
- [x] **No**
|
| 114 |
+
|
| 115 |
+
If yes, describe: `TBD`
|
| 116 |
+
|
| 117 |
+
---
|
| 118 |
+
|
| 119 |
+
## 💡 Diversity Dimensions
|
| 120 |
+
|
| 121 |
+
- [ ] **Camera Position / Angle**
|
| 122 |
+
- [ ] **Lighting Conditions**
|
| 123 |
+
- [ ] **Target Object**
|
| 124 |
+
- [ ] **Spatial Layout**
|
| 125 |
+
- [ ] **Robot Embodiment**
|
| 126 |
+
- [ ] **Task Execution**
|
| 127 |
+
- [ ] **Background / Scene**
|
| 128 |
+
- [ ] **Other** (Please specify: `TBD`)
|
| 129 |
+
|
| 130 |
+
Elaboration: `TBD`
|
| 131 |
+
|
| 132 |
+
---
|
| 133 |
+
|
| 134 |
+
## 🛠️ Equipment & Setup
|
| 135 |
+
|
| 136 |
+
|
| 137 |
+
### Sensors & Cameras
|
| 138 |
+
|
| 139 |
+
| Type | Model/Details |
|
| 140 |
+
| :--- | :--- |
|
| 141 |
+
| **Primary Camera** | `Monocular RGB video (resolution in meta/episodes_stats.jsonl)` |
|
| 142 |
+
| **Depth** | `Per-frame depth stored as NPZ (raw/depth) + depth_vis mp4 (videos/...)` |
|
| 143 |
+
| **Segmentation** | `Part/label-map PNGs (raw/seg) + part_id mp4 (videos/...)` |
|
| 144 |
+
| **Other** | `Dynamic 3D Gaussians (optional) in raw/gaussians/` |
|
| 145 |
+
|
| 146 |
+
---
|
| 147 |
+
|
| 148 |
+
## 🎯 Action & State Space Representation
|
| 149 |
+
|
| 150 |
+
This dataset is primarily **vision + tool-state** oriented. Robot kinematics may not be included.
|
| 151 |
+
|
| 152 |
+
### Action Space Representation
|
| 153 |
+
|
| 154 |
+
**Primary Action Representation:**
|
| 155 |
+
- [ ] **Absolute Cartesian**
|
| 156 |
+
- [ ] **Relative Cartesian**
|
| 157 |
+
- [ ] **Joint Space**
|
| 158 |
+
- [x] **Other**: *Action is a categorical label per episode (`action.label`)*
|
| 159 |
+
|
| 160 |
+
**Action Dimensions:**
|
| 161 |
+
```
|
| 162 |
+
action.label: string
|
| 163 |
+
- Categorical action label (same for all timesteps in an episode)
|
| 164 |
+
```
|
| 165 |
+
|
| 166 |
+
### State Space Representation
|
| 167 |
+
|
| 168 |
+
**State Information Included:**
|
| 169 |
+
- [ ] **Joint Positions**
|
| 170 |
+
- [ ] **Joint Velocities**
|
| 171 |
+
- [ ] **End-Effector Pose**
|
| 172 |
+
- [ ] **Force/Torque Readings**
|
| 173 |
+
- [ ] **Gripper State**
|
| 174 |
+
- [x] **Other**: *2D tool pose state (per tool)*
|
| 175 |
+
|
| 176 |
+
**State Dimensions:**
|
| 177 |
+
```
|
| 178 |
+
observation.state.tool_pose_2d: [num_tools, 3]
|
| 179 |
+
- tx_px: tool center / tip x position in pixels (see toolposes JSON definition)
|
| 180 |
+
- ty_px: tool center / tip y position in pixels
|
| 181 |
+
- theta_rad: in-plane rotation (radians)
|
| 182 |
+
```
|
| 183 |
+
|
| 184 |
+
Additional related fields in parquet:
|
| 185 |
+
- `observation.meta.tool_name`: list of tool names (ordering reference)
|
| 186 |
+
- `observation.meta.tool_visible`: per-tool boolean mask
|
| 187 |
+
- `observation.meta.depth_npz_relpath`: pointer to raw depth NPZ per timestep
|
| 188 |
+
- `observation.meta.seg_png_relpath`: pointer to raw part label-map PNG per timestep
|
| 189 |
+
|
| 190 |
+
### Dynamic 3D Gaussians (Optional)
|
| 191 |
+
|
| 192 |
+
If present, per-episode files live under:
|
| 193 |
+
- `raw/gaussians/episode_XXXXXX/dynamic_gaussians_frames.jsonl.gz`
|
| 194 |
+
- `raw/gaussians/episode_XXXXXX/dynamic_gaussians_meta.json`
|
| 195 |
+
- `raw/gaussians/episode_XXXXXX/photometric_cam_T_wc.json`
|
| 196 |
+
|
| 197 |
+
These encode per-frame 3D Gaussian parameters (positions/covariances/colors/opacities, etc. as reconstructed by MoSca https://github.com/JiahuiLei/MoSca) as benchtop scene representations and camera transforms (intrinsics and extrinsics) for 3D reconstruction or tracking research.
|
| 198 |
+
|
| 199 |
+
---
|
| 200 |
+
|
| 201 |
+
## ⏱️ Data Synchronization Approach
|
| 202 |
+
|
| 203 |
+
All modalities are aligned **by frame index** during export.
|
| 204 |
+
|
| 205 |
+
- The RGB video provides the reference frame count.
|
| 206 |
+
- Depth NPZ frames, segmentation PNG frames, and tool pose frames are truncated to:
|
| 207 |
+
```
|
| 208 |
+
n_steps = min(rgb_frames, depth_npz_count, seg_png_count, toolposes_num_frames)
|
| 209 |
+
```
|
| 210 |
+
- Timestamps are generated as:
|
| 211 |
+
```
|
| 212 |
+
timestamp_s = step_index / fps_timestamps
|
| 213 |
+
```
|
| 214 |
+
where `fps_timestamps` is recorded in `meta/info.json`.
|
| 215 |
+
|
| 216 |
+
This approach ensures consistent timestep alignment across all exported modalities.
|
| 217 |
+
|
| 218 |
+
---
|
| 219 |
+
|
| 220 |
+
## 👥 Attribution & Contact
|
| 221 |
+
|
| 222 |
+
| | |
|
| 223 |
+
| :--- | :--- |
|
| 224 |
+
| **Dataset Lead** | Nan Xiao, Tom Olesch, Farong Wang, Fei Liu, DeLong Jonathan C. |
|
| 225 |
+
| **Institution** | University of Tennessee, Knoxville (AURAS Lab) |
|
| 226 |
+
| **Contact Email** | nxiao4@vols.utk.edu |
|
| 227 |
+
| **Citation (BibTeX)** | <pre><code>@misc{benchtop_openh_2026,<br> author = {Xiao, Nan and collaborators},<br> title = {Benchtop Surgical Tool Manipulation Dataset (Open-H)},<br> year = {2026},<br> publisher = {Open-H-Embodiment},<br> note = {Dataset packaging in LeRobot v2.1-style layout}<br>}</code></pre> |
|
Surgical/utenn/benchtop_datasets_round2_with_part_seg/meta/episodes.jsonl
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{"episode_index": 0, "episode_id": "episode_000000", "task": "PULLING", "clip_id": "t2_video", "fps_timestamps": 10.0, "fps_videos": 10.0, "num_steps": 127, "duration_s": 12.6, "videos": {"observation.images.rgb": "videos/chunk-000/observation.images.rgb/episode_000000.mp4", "observation.images.depth_vis": "videos/chunk-000/observation.images.depth_vis/episode_000000.mp4", "observation.images.part_id": "videos/chunk-000/observation.images.part_id/episode_000000.mp4"}, "parquet": "data/chunk-000/episode_000000.parquet", "raw_paths": {"raw_toolposes_dir": "raw/toolposes/episode_000000", "raw_depth_dir": "raw/depth/episode_000000", "raw_seg_dir": "raw/seg/episode_000000", "raw_seg_schema_dir": "raw/seg_schema/episode_000000", "raw_gaussians_dir": "raw/gaussians/episode_000000"}, "tools": {"tool_names": ["grasper", "grasper2"], "tool_name_to_slot": {"grasper": 0, "grasper2": 1}, "pose_definition": "[tx_px, ty_px, theta_rad] per tool; NaN if not visible"}, "modalities_present": {"depth": true, "segmentation": true, "dynamic_gaussians": true}, "dynamic_gaussians": {"present": true, "frames_jsonl_gz": "raw/gaussians/episode_000000/dynamic_gaussians_frames.jsonl.gz", "meta_json": "raw/gaussians/episode_000000/dynamic_gaussians_meta.json", "cam_T_wc_json": "raw/gaussians/episode_000000/photometric_cam_T_wc.json"}}
|
| 2 |
+
{"episode_index": 1, "episode_id": "episode_000001", "task": "PULLING", "clip_id": "t4_video", "fps_timestamps": 10.0, "fps_videos": 10.0, "num_steps": 96, "duration_s": 9.5, "videos": {"observation.images.rgb": "videos/chunk-000/observation.images.rgb/episode_000001.mp4", "observation.images.depth_vis": "videos/chunk-000/observation.images.depth_vis/episode_000001.mp4", "observation.images.part_id": "videos/chunk-000/observation.images.part_id/episode_000001.mp4"}, "parquet": "data/chunk-000/episode_000001.parquet", "raw_paths": {"raw_toolposes_dir": "raw/toolposes/episode_000001", "raw_depth_dir": "raw/depth/episode_000001", "raw_seg_dir": "raw/seg/episode_000001", "raw_seg_schema_dir": "raw/seg_schema/episode_000001", "raw_gaussians_dir": "raw/gaussians/episode_000001"}, "tools": {"tool_names": ["grasper", "grasper2"], "tool_name_to_slot": {"grasper": 0, "grasper2": 1}, "pose_definition": "[tx_px, ty_px, theta_rad] per tool; NaN if not visible"}, "modalities_present": {"depth": true, "segmentation": true, "dynamic_gaussians": true}, "dynamic_gaussians": {"present": true, "frames_jsonl_gz": "raw/gaussians/episode_000001/dynamic_gaussians_frames.jsonl.gz", "meta_json": "raw/gaussians/episode_000001/dynamic_gaussians_meta.json", "cam_T_wc_json": "raw/gaussians/episode_000001/photometric_cam_T_wc.json"}}
|
Surgical/utenn/benchtop_datasets_round2_with_part_seg/meta/episodes_stats.jsonl
ADDED
|
@@ -0,0 +1,2 @@
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{"episode_index": 0, "action": "PULLING", "clip_id": "t2_video", "num_steps": 127, "duration_s": 12.6, "fps_timestamps": 10.0, "fps_videos": 10.0, "rgb_hw": [416, 720], "num_tools": 2, "tool_names": ["grasper", "grasper2"], "depth_present": true, "seg_present": true, "gaussians_present": true}
|
| 2 |
+
{"episode_index": 1, "action": "PULLING", "clip_id": "t4_video", "num_steps": 96, "duration_s": 9.5, "fps_timestamps": 10.0, "fps_videos": 10.0, "rgb_hw": [416, 720], "num_tools": 2, "tool_names": ["grasper", "grasper2"], "depth_present": true, "seg_present": true, "gaussians_present": true}
|
Surgical/utenn/benchtop_datasets_round2_with_part_seg/meta/info.json
ADDED
|
@@ -0,0 +1,75 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"repo_id": "utk/benchtop_v1",
|
| 3 |
+
"dataset_format_version": "2.1",
|
| 4 |
+
"lerobot_version_target": "0.3.3",
|
| 5 |
+
"license": "CC BY 4.0",
|
| 6 |
+
"robot_type": "unknown",
|
| 7 |
+
"use_videos": true,
|
| 8 |
+
"fps_timestamps": 10.0,
|
| 9 |
+
"fps_videos": 10.0,
|
| 10 |
+
"modalities": [
|
| 11 |
+
"observation.images.rgb",
|
| 12 |
+
"observation.images.depth_vis",
|
| 13 |
+
"observation.images.part_id"
|
| 14 |
+
],
|
| 15 |
+
"additional_fields": {
|
| 16 |
+
"observation.meta.tool_name": {
|
| 17 |
+
"dtype": "string",
|
| 18 |
+
"shape": [
|
| 19 |
+
"num_tools"
|
| 20 |
+
],
|
| 21 |
+
"description": "Tool name list; defines ordering for tool-visible and 2D pose arrays."
|
| 22 |
+
},
|
| 23 |
+
"observation.meta.tool_visible": {
|
| 24 |
+
"dtype": "bool",
|
| 25 |
+
"shape": [
|
| 26 |
+
"num_tools"
|
| 27 |
+
],
|
| 28 |
+
"description": "Per-tool visibility flag for the 2D pose."
|
| 29 |
+
},
|
| 30 |
+
"observation.state.tool_pose_2d": {
|
| 31 |
+
"dtype": "float32",
|
| 32 |
+
"shape": [
|
| 33 |
+
"num_tools",
|
| 34 |
+
3
|
| 35 |
+
],
|
| 36 |
+
"names": [
|
| 37 |
+
"tx_px",
|
| 38 |
+
"ty_px",
|
| 39 |
+
"theta_rad"
|
| 40 |
+
],
|
| 41 |
+
"description": "Per-tool 2D pose [tx,ty,theta] aligned with tool_name ordering. NaN if not visible."
|
| 42 |
+
},
|
| 43 |
+
"observation.meta.depth_npz_relpath": {
|
| 44 |
+
"dtype": "string",
|
| 45 |
+
"shape": [
|
| 46 |
+
1
|
| 47 |
+
],
|
| 48 |
+
"description": "Relative path to raw depth .npz under raw/depth/episode_xxxxxx/ (only present when depth exists)."
|
| 49 |
+
},
|
| 50 |
+
"observation.meta.seg_png_relpath": {
|
| 51 |
+
"dtype": "string",
|
| 52 |
+
"shape": [
|
| 53 |
+
1
|
| 54 |
+
],
|
| 55 |
+
"description": "Relative path to raw label_map png under raw/seg/episode_xxxxxx/ (only present when segmentation exists)."
|
| 56 |
+
}
|
| 57 |
+
},
|
| 58 |
+
"splits": {
|
| 59 |
+
"train": "0:1",
|
| 60 |
+
"val": "1:1",
|
| 61 |
+
"test": "1:2"
|
| 62 |
+
},
|
| 63 |
+
"num_episodes": 2,
|
| 64 |
+
"chunking": {
|
| 65 |
+
"chunks": [
|
| 66 |
+
"chunk-000"
|
| 67 |
+
]
|
| 68 |
+
},
|
| 69 |
+
"notes": {
|
| 70 |
+
"truncation_policy": "n_steps = min(counts over available modalities: rgb, toolposes, and optionally depth/seg if present)",
|
| 71 |
+
"depth_storage": "raw npz per frame + depth_vis video for viewing (only if depth is available)",
|
| 72 |
+
"seg_storage": "raw label_map png per frame + part_id video for viewing (only if segmentation is available)",
|
| 73 |
+
"gaussians_storage": "episode-level raw copy if present"
|
| 74 |
+
}
|
| 75 |
+
}
|
Surgical/utenn/benchtop_datasets_round2_with_part_seg/meta/tasks.jsonl
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"task": "PULLING", "description": "Benchtop action label: PULLING"}
|
Surgical/utenn/surgical_video_datasets_round2_with_part_seg/videos/chunk-000/observation.images.tool_segmentation_id/episode_000005.mkv
ADDED
|
Binary file (76.2 kB). View file
|
|
|
Surgical/utenn/surgical_video_datasets_round2_with_part_seg/videos/chunk-000/observation.images.tool_segmentation_id/episode_000015.mkv
ADDED
|
Binary file (83.5 kB). View file
|
|
|
Surgical/utenn/surgical_video_datasets_round2_with_part_seg/videos/chunk-000/observation.images.tool_segmentation_id/episode_000029.mkv
ADDED
|
Binary file (73 kB). View file
|
|
|
Surgical/utenn/surgical_video_datasets_round2_with_part_seg/videos/chunk-000/observation.images.tool_segmentation_id/episode_000039.mkv
ADDED
|
Binary file (92.9 kB). View file
|
|
|
Surgical/utenn/surgical_video_datasets_round2_with_part_seg/videos/chunk-000/observation.images.tool_segmentation_id/episode_000056.mkv
ADDED
|
Binary file (68.8 kB). View file
|
|
|
Surgical/utenn/surgical_video_datasets_round2_with_part_seg/videos/chunk-000/observation.images.tool_segmentation_id/episode_000073.mkv
ADDED
|
Binary file (43.1 kB). View file
|
|
|
Surgical/utenn/surgical_video_datasets_round2_with_part_seg/videos/chunk-000/observation.images.tool_segmentation_id/episode_000077.mkv
ADDED
|
Binary file (92.4 kB). View file
|
|
|
Surgical/utenn/surgical_video_datasets_round2_with_part_seg/videos/chunk-000/observation.images.tool_segmentation_id/episode_000078.mkv
ADDED
|
Binary file (27.2 kB). View file
|
|
|
Surgical/utenn/surgical_video_datasets_round2_with_part_seg/videos/chunk-000/observation.images.tool_segmentation_id/episode_000080.mkv
ADDED
|
Binary file (24.7 kB). View file
|
|
|
Surgical/utenn/surgical_video_datasets_round2_with_part_seg/videos/chunk-000/observation.images.tool_segmentation_id/episode_000083.mkv
ADDED
|
Binary file (42.8 kB). View file
|
|
|
Surgical/utenn/surgical_video_datasets_round2_with_part_seg/videos/chunk-000/observation.images.tool_segmentation_id/episode_000084.mkv
ADDED
|
Binary file (59 kB). View file
|
|
|
Surgical/utenn/surgical_video_datasets_round2_with_part_seg/videos/chunk-000/observation.images.tool_segmentation_id/episode_000093.mkv
ADDED
|
Binary file (58.6 kB). View file
|
|
|
Surgical/utenn/surgical_video_datasets_round2_with_part_seg/videos/chunk-000/observation.images.tool_segmentation_id/episode_000094.mkv
ADDED
|
Binary file (28 kB). View file
|
|
|
Surgical/utenn/surgical_video_datasets_round2_with_part_seg/videos/chunk-000/observation.images.tool_segmentation_id/episode_000098.mkv
ADDED
|
Binary file (55.1 kB). View file
|
|
|
Surgical/utenn/surgical_video_datasets_round2_with_part_seg/videos/chunk-000/observation.images.tool_segmentation_id/episode_000100.mkv
ADDED
|
Binary file (47.4 kB). View file
|
|
|