OpenParable
Skip to content
← Back to skills

Depth Estimation

DevOps & Security
depth-estimation

Depth Estimation is an Agent Skill for devops & security published in the SharpAI/DeepCamera repository on GitHub: a SKILL.md instruction file an AI agent can load to do this task.

Real-time depth map privacy transforms using Depth Anything v2 (CoreML + PyTorch)

View on GitHub ↗
QUICK START

How to use this skill

Bring this guide into your coding agent with a prompt tailored to the tool you use.

  1. Open your project in Codex.
  2. Copy and paste the prompt below.
  3. Review files and risks before approving any action.
Prompt to paste
Inspect the Agent Skill at https://github.com/SharpAI/DeepCamera/blob/HEAD/SKILL.md and all supporting files. Report provenance, permissions, scripts, dependencies, and security risks. Propose an exact installation into .agents/skills/depth-estimation, but wait for my approval before writing files or running scripts.

Copying this prompt does not install or run the skill. Review third-party files before use.

Depth Estimation (Privacy)

Real-time monocular depth estimation using Depth Anything v2. Transforms camera feeds with colorized depth maps — near objects appear warm, far objects appear cool.

When used for privacy mode, the depth_only blend mode fully anonymizes the scene while preserving spatial layout and activity, enabling security monitoring without revealing identities.

Hardware Backends

PlatformBackendRuntimeModel
macOSCoreMLApple Neural Engineapple/coreml-depth-anything-v2-small (.mlpackage)
Linux/WindowsPyTorchCUDA / CPUdepth-anything/Depth-Anything-V2-Small (.pth)

On macOS, CoreML runs on the Neural Engine, leaving the GPU free for other tasks. The model is auto-downloaded from HuggingFace and stored at ~/.aegis-ai/models/feature-extraction/.

What You Get

  • Privacy anonymization — depth-only mode hides all visual identity
  • Depth overlays on live camera feeds
  • 3D scene understanding — spatial layout of the scene
  • CoreML acceleration — Neural Engine on Apple Silicon (3-5x faster than MPS)

Interface: TransformSkillBase

This skill implements the TransformSkillBase interface. Any new privacy skill can be created by subclassing TransformSkillBase and implementing two methods:

from transform_base import TransformSkillBase

class MyPrivacySkill(TransformSkillBase):
    def load_model(self, config):
        # Load your model, return {"model": "...", "device": "..."}
        ...

    def transform_frame(self, image, metadata):
        # Transform BGR image, return BGR image
        ...

Protocol

Aegis → Skill (stdin)

{"event": "frame", "frame_id": "cam1_1710001", "camera_id": "front_door", "frame_path": "/tmp/frame.jpg", "timestamp": "..."}
{"command": "config-update", "config": {"opacity": 0.8, "blend_mode": "overlay"}}
{"command": "stop"}

Skill → Aegis (stdout)

{"event": "ready", "model": "coreml-DepthAnythingV2SmallF16", "device": "neural_engine", "backend": "coreml"}
{"event": "transform", "frame_id": "cam1_1710001", "camera_id": "front_door", "transform_data": "<base64 JPEG>"}
{"event": "perf_stats", "total_frames": 50, "timings_ms": {"transform": {"avg": 12.5, ...}}}

Setup

python3 -m venv .venv && source .venv/bin/activate
pip install -r requirements.txt