skills.rest
atomicrajat
atomicrajatCOMMUNITY
Explore all skills by this author →
GitHub

vss-deploy-detection-tracking-2d

Deploy, debug, and operate the RTVI-CV 2D detection and tracking microservice via Docker and REST API.

2|Updated Aug 20, 2026
Search Tags:#deployment#docker#rest api#object detection#nvidia#video analytics#deepstream
One-click install
npx skills add https://github.com/atomicrajat/industry_safety_monitoring_system --skill vss-deploy-detection-tracking-2d-atomicrajat
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: vss-deploy-detection-tracking-2d
Source: https://github.com/atomicrajat/industry_safety_monitoring_system/tree/main/.claude/skills/vss-deploy-detection-tracking-2d
Command: npx skills add https://github.com/atomicrajat/industry_safety_monitoring_system --skill vss-deploy-detection-tracking-2d-atomicrajat

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve? Deploying and operating NVIDIA's RTVI-CV 2D detection/tracking microservice involves many error-prone steps: pulling NGC resources, resolving model and video paths, editing DeepStream configs, managing TensorRT engine caches, and driving the REST API for streams, health, and metrics. This Skill turns that into a guided, step-by-step workflow with deterministic helper scripts. ## Core Features & Use Cases - Guided deployment: End-to-end deploy flow for warehouse-2d, warehouse-3d, smartcity-rtdetr, and smartcity-gdino use cases, with platform detection (x86 dGPU, SBSA, Jetson) and per-step status boxes. - REST API operations: Add/remove/list streams, probe liveness/readiness/startup, collect FPS and GPU metrics, and generate text embeddings against a running instance on port 9000. - Debug and teardown: Troubleshooting runbooks for healthcheck failures, NGC auth errors, and GPU OOM, plus clean container teardown flows. - Use Case: An engineer says "deploy rtvi-cv warehouse 2d with 4 streams and display" — the Skill resolves the ONNX model and videos, applies batch/sink/source configuration inside the container, waits for readiness, and reports live FPS. ## Quick Start Ask the agent to deploy rtvi-cv warehouse 2d with 4 streams and display, then follow the step-by-step prompts it presents.

Frequently Asked Questions about vss-deploy-detection-tracking-2d

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I deploy the RTVI-CV 2D detection microservice?▼

Ask to deploy rtvi-cv with a use case such as warehouse-2d, smartcity-rtdetr, or smartcity-gdino. The Skill resolves NGC resources, applies DeepStream configuration inside the container, launches the perception app, and waits for readiness before reporting metrics.

How do I add or remove streams on a running rtvi-cv instance?▼

Use the REST API on port 9000: POST /api/v1/stream/add or /stream/remove with a camera_id and camera_url. The add_streams.sh helper adds streams one at a time with a delay to avoid caps-negotiation stalls in DeepStream.

What hardware does the RTVI-CV microservice support?▼

It runs on x86 and aarch64 dGPU systems (T4, A100, L40, H100, B200, RTX), SBSA platforms like Grace-Hopper, and Jetson devices including Thor, Orin, and Xavier. The Skill detects the platform and resolves matching defaults.

Why does the rtvi-cv container fail with NGC 401 or 403 errors?▼

HTTP 401/403 during image or resource pulls means the NGC_CLI_API_KEY is missing or expired. Run docker login nvcr.io and re-export the key before retrying the deployment.

What are the limitations of this deployment skill?▼

It requires the matching VSS profile to be deployed and reachable, plus NGC credentials for pulls. NGC-hosted models may have rate limits, GPU memory requirements, and license restrictions, and concurrency limits depend on host hardware.

Related Skills

Explore More Software Engineering Skills→
NVIDIANVIDIA
3.2k

vss-deploy-detection-tracking-3d

Deploy the RTVI-CV-3D MV3DT microservice for multi-camera 3D detection and tracking.

Official
Advanced
NVIDIA-AI-BlueprintsNVIDIA-AI-Blueprints
1.8k

rt-vlm

Deploys the VLM service for real-time video captioning and incident detection.

Official
Intermediate
NVIDIA-AI-BlueprintsNVIDIA-AI-Blueprints
1.8k

vss-deploy-detection-tracking-3d

Deploy the RTVI-CV-3D MV3DT microservice for multi-camera 3D detection and tracking.

Official
Advanced
NVIDIANVIDIA
3.2k

deepstream-import-vision-model

Import object detection models from HuggingFace or NGC into NVIDIA DeepStream pipelines with automated benchmarking.

Official
Advanced