Provide real-time per-pixel depth frames from Spectacles via DepthModule.
npx skills add https://github.com/specs-devs/agentic-tools --skill api-depth
Please help me install this Agent Skill. Skill: api-depth Source: https://github.com/specs-devs/agentic-tools/tree/main/skills/api-depth Command: npx skills add https://github.com/specs-devs/agentic-tools --skill api-depth
Provide real-time depth frames from Spectacles to enable depth-based AR placement, occlusion, and spatial understanding.
Initialize the DepthModule and start a depth frame session to access per-pixel depth data in world space.
High-intent search queries and answers about installing and using this skill.
Real-time depth data for AR occlusion is provided through per-pixel depth frames from the DepthModule. You can use this data to handle scene occlusion and place AR objects accurately in Lens Studio projects.
Yes, you can convert depth frames to 3D world space by back-projecting per-pixel depth values using camera intrinsics. This enables accurate spatial mapping and real-time object placement in your AR scene.
To start a depth frame session for spatial mapping on Spectacles, initialize the DepthModule. This exposes DepthFrameData, providing the real-time depth values needed for environmental understanding.
Yes, camera access is required to use the depth module. Because the feature is experimental, accessing real-time per-pixel depth data and camera intrinsics depends directly on camera availability.
Limitations of using real-time depth data include its experimental status and the requirement for camera access. It provides per-pixel depth and back-projection to world space, which may have constraints in complex spatial mapping environments.
specs-devsAccess live camera frames, still images, and intrinsics via Lens Studio CameraModule.
CUHK-AIM-GroupTranslate neuroimaging outputs into publication-ready figures and 3D meshes.
specs-devsRun on-device ML inference for Spectacles with SnapML on camera frames.
specs-devsPersist AR content at fixed world positions using the Spatial Anchors API.
