lingbot-depth
Robbyant
Masked Depth Modeling for Spatial Perception
perception
Verify weight files, sizes, format, and loading instructions.
A model hub link is a declaration. Files, loadability, evaluation, and deployment are scored separately.
Model decision scorecard
Use-case scores and evidence confidence are separate; Unknown is not treated as failure.
Access and governance
UsefulAccess and license are declared by the source.
Artifact availability
UnknownCode and weights are not verified.
Training and loading reproducibility
UnknownNo verified loading configuration is available.
Training data requirements
UnknownTraining data requirements are not structured enough for reliable dataset matching.
Evaluation evidence
UnknownNo structured evaluation evidence has been verified.
Deployment readiness
UnknownHardware, latency, dependencies, and checkpoint loading are not pipeline-tested.
Artifact facts and provenance
availabilitymetadata_verifiedopen
repository metadata
licensemetadata_verifiedApache-2.0
repository metadata: license
dataset.signalsofficial_claimdepth
README/card declaration
Loop signal demand
Signals this model family needs for training, evaluation, or failure mining.
Observation / ego video
depth · depth · Metric depth accuracy · Depth completion
Language intent / task phase
Needs task intent, instructions, keysteps, or phase labels to connect behavior to goals.
Action / robot state
Needs executable actions, robot state, hand pose, or trajectory supervision.
Future state / dynamics
Needs future-state supervision or rollout structure to validate predictive dynamics.
Feedback / correction / failure
Needs success, failure, correction, or recovery signals to turn evaluation into better data.
Sim-real / embodiment metadata
depth · depth · Metric depth accuracy · Depth completion
Evaluation focus
- Metric depth accuracy
- Depth completion
- 4D point tracking
- Geometry-sensitive grasping
Missing critical loop signals
Core signal demands are represented. Check quality, alignment, and access constraints.
Related catalog datasets
RH20T
Contact-rich multimodal manipulation paired with human demonstrations
Dataset license restricts commercial use.
DROID
In-the-wild real-world robot manipulation
Exact action dimensions, control frequency, normalization, and camera mapping require interface verification.
Ego-Exo4D
Synchronized first-person and third-person skilled activity
Exact action dimensions, control frequency, normalization, and camera mapping require interface verification.
OpenBot notes
- The recommended v0.5 release supersedes the earlier v0.1 checkpoint.
- Training includes RobbyReal, RobbyVla, and RobbySim subsets.