{
  "version": "https://jsonfeed.org/version/1.1",
  "title": "Khaled Sharif - Notes",
  "description": "Technical notes and articles on computer vision, robotics, and machine learning",
  "home_page_url": "https://kldsrf.com/notes",
  "feed_url": "https://kldsrf.com/feed.json",
  "language": "en",
  "icon": "https://kldsrf.com/favicon.ico",
  "authors": [
    {
      "name": "Khaled Sharif",
      "url": "https://kldsrf.com"
    }
  ],
  "items": [
    {
      "id": "https://kldsrf.com/notes/camera-calibration",
      "url": "https://kldsrf.com/notes/camera-calibration",
      "title": "Camera Calibration",
      "content_text": "Zhang's method, distortion models, stereo and hand-eye calibration, and online refinement for production perception systems.",
      "summary": "Zhang's method, distortion models, stereo and hand-eye calibration, and online refinement for production perception systems.",
      "tags": [
        "camera calibration",
        "computer vision",
        "perception",
        "distortion",
        "stereo vision",
        "production systems"
      ],
      "_emoji": "📷"
    },
    {
      "id": "https://kldsrf.com/notes/classical-vs-modern-cv",
      "url": "https://kldsrf.com/notes/classical-vs-modern-cv",
      "title": "Classical vs Modern CV",
      "content_text": "When geometric methods beat deep learning in production perception—and hybrid patterns that combine both.",
      "summary": "When geometric methods beat deep learning in production perception—and hybrid patterns that combine both.",
      "tags": [
        "computer vision",
        "deep learning",
        "classical cv",
        "geometric reasoning",
        "hybrid systems",
        "production systems"
      ],
      "_emoji": "⚖️"
    },
    {
      "id": "https://kldsrf.com/notes/deploying-cv-production",
      "url": "https://kldsrf.com/notes/deploying-cv-production",
      "title": "Deploying CV in Production",
      "content_text": "Optimization, edge and cloud deployment, reliability engineering, and monitoring for production computer vision.",
      "summary": "Optimization, edge and cloud deployment, reliability engineering, and monitoring for production computer vision.",
      "tags": [
        "production systems",
        "deployment",
        "tensorrt",
        "quantization",
        "monitoring",
        "computer vision"
      ],
      "_emoji": "🚀"
    },
    {
      "id": "https://kldsrf.com/notes/formal-verification",
      "url": "https://kldsrf.com/notes/formal-verification",
      "title": "Formal Verification",
      "content_text": "Model checking, theorem proving, and NN robustness methods used for ISO 26262 and SOTIF automotive safety cases.",
      "summary": "Model checking, theorem proving, and NN robustness methods used for ISO 26262 and SOTIF automotive safety cases.",
      "tags": [
        "formal verification",
        "automotive",
        "safety",
        "model checking",
        "theorem proving",
        "sotif"
      ],
      "_emoji": "✅"
    },
    {
      "id": "https://kldsrf.com/notes/collision-avoidance",
      "url": "https://kldsrf.com/notes/collision-avoidance",
      "title": "Collision Avoidance",
      "content_text": "Perception, planning, sensor fusion, V2X, and safety models (RSS) for autonomous-vehicle collision avoidance.",
      "summary": "Perception, planning, sensor fusion, V2X, and safety models (RSS) for autonomous-vehicle collision avoidance.",
      "tags": [
        "collision avoidance",
        "autonomous vehicles",
        "path planning",
        "obstacle detection",
        "sensors",
        "safety"
      ],
      "_emoji": "🚧"
    },
    {
      "id": "https://kldsrf.com/notes/sensor-fusion",
      "url": "https://kldsrf.com/notes/sensor-fusion",
      "title": "Sensor Fusion",
      "content_text": "Camera–LiDAR–radar fusion architectures, industry stacks, and BEV/transformer trends for AV perception.",
      "summary": "Camera–LiDAR–radar fusion architectures, industry stacks, and BEV/transformer trends for AV perception.",
      "tags": [
        "sensor fusion",
        "autonomous vehicles",
        "perception",
        "lidar",
        "radar",
        "kalman filter"
      ],
      "_emoji": "🤖"
    },
    {
      "id": "https://kldsrf.com/notes/slam-sota",
      "url": "https://kldsrf.com/notes/slam-sota",
      "title": "Dense SLAM",
      "content_text": "Dense SLAM with Gaussian splatting, monocular sparse tracking, and anti-dynamics pipelines for real-time mapping.",
      "summary": "Dense SLAM with Gaussian splatting, monocular sparse tracking, and anti-dynamics pipelines for real-time mapping.",
      "tags": [
        "slam",
        "localization",
        "mapping",
        "gaussian splatting",
        "research",
        "computer vision"
      ],
      "_emoji": "📖"
    },
    {
      "id": "https://kldsrf.com/notes/auto-nav",
      "url": "https://kldsrf.com/notes/auto-nav",
      "title": "Autonomous Navigation",
      "content_text": "LiDAR place recognition, NeRF driving maps, anomaly detection, and qualitative SLAM for autonomous robot navigation.",
      "summary": "LiDAR place recognition, NeRF driving maps, anomaly detection, and qualitative SLAM for autonomous robot navigation.",
      "tags": [
        "autonomous navigation",
        "robotics",
        "path planning",
        "place recognition",
        "slam"
      ],
      "_emoji": "🧭"
    },
    {
      "id": "https://kldsrf.com/notes/bev",
      "url": "https://kldsrf.com/notes/bev",
      "title": "Bird's Eye View",
      "content_text": "Top-down BEV perception for AVs: multi-camera fusion, place recognition, segmentation, and unified detection–prediction.",
      "summary": "Top-down BEV perception for AVs: multi-camera fusion, place recognition, segmentation, and unified detection–prediction.",
      "tags": [
        "bev",
        "birds eye view",
        "perception",
        "sensor fusion",
        "autonomous driving",
        "computer vision"
      ],
      "_emoji": "🐦"
    },
    {
      "id": "https://kldsrf.com/notes/nerfs",
      "url": "https://kldsrf.com/notes/nerfs",
      "title": "Neural Radiance Fields",
      "content_text": "NeRFs for robotics: novel-view synthesis, localization on neural maps, dynamic SLAM, and real-time object mapping.",
      "summary": "NeRFs for robotics: novel-view synthesis, localization on neural maps, dynamic SLAM, and real-time object mapping.",
      "tags": [
        "nerf",
        "neural radiance fields",
        "3d reconstruction",
        "novel view synthesis",
        "robotics",
        "slam"
      ],
      "_emoji": "🪐"
    },
    {
      "id": "https://kldsrf.com/notes/vision-language-action",
      "url": "https://kldsrf.com/notes/vision-language-action",
      "title": "Vision-Language-Action",
      "content_text": "VLA landscape from RT-2 to OpenPI π₀.₅: co-training recipes, hierarchical reasoning, and open-world robot control.",
      "summary": "VLA landscape from RT-2 to OpenPI π₀.₅: co-training recipes, hierarchical reasoning, and open-world robot control.",
      "tags": [
        "vla",
        "vision language action",
        "robotics",
        "foundation models",
        "openpi",
        "embodied ai"
      ],
      "_emoji": "🦾"
    }
  ]
}