NuRec Web Viewer

Web-based viewer for NuRec neural reconstruction scenarios. Design is consistent with the PhysicalAI-AV Camera Replay webapp.

Features

• Multi-Camera Display: View up to 3 cameras simultaneously with draggable windows
• Layout Presets: Grid, Surround (vehicle perspective), Stack layouts
• Playback Controls: Play/Pause, ±1s/±5s step, timeline scrubbing
• Speed Control: 0.5x, 1x, 2x playback speed
• Quality Control: Low/Medium/High rendering quality
• Camera Selection: Choose which cameras to display
• Status Bar: Connection status and renderer info

Architecture

Frontend (React)  -->  Backend (FastAPI)  -->  NuRec Container (gRPC)
   :3000                  :8000                     :46435

Prerequisites

1. NuRec Container Image: Either:
   • Docker image: NUREC_IMAGE=carlasimulator/nvidia-nurec-grpc:0.2.0
2. USDZ Scene File: A valid NuRec scene file (download from HuggingFace)
3. GPU: NVIDIA GPU with CUDA support
4. Node.js: v18+ for frontend
5. Python: 3.10+ for backend

Quick Start

1. Download USDZ Scene

# Login first (token must have gated dataset read access)
hf auth login

# Download one USDZ file to current directory
hf download nvidia/PhysicalAI-Autonomous-Vehicles-NuRec \
  "sample_set/25.07_release/Batch0001/026d6a39-bd8f-4175-bc61-fe50ed0403a3/026d6a39-bd8f-4175-bc61-fe50ed0403a3.usdz" \
  --repo-type dataset \
  --local-dir .

2. Start NuRec Container

export NUREC_IMAGE=carlasimulator/nvidia-nurec-grpc:0.2.0
./scripts/start_nurec.sh /path/to/scene.usdz 46435

Wait for the message "successfully loaded scene" or "serving on localhost:46435".

3. Start Backend

cd backend
pip install -r requirements.txt
./run.sh

Backend will be available at http://localhost:8000

4. Start Frontend

cd frontend
npm install
npm run dev

Frontend will be available at http://localhost:3000

5. Load Scene

1. Open http://localhost:3000 in your browser
2. Enter the USDZ file path
3. Click "Load"
4. Select cameras and use playback controls