@camstack/addon-vision 0.1.2 → 0.1.3

package/dist/addons/object-detection/index.mjs

@@ -1,11 +1,377 @@
 import {
-  ObjectDetectionAddon
-} from "../../chunk-R5J3WAUI.mjs";
-import "../../chunk-KUO2BVFY.mjs";
-import "../../chunk-BP7H4NFS.mjs";
-import "../../chunk-22BHCDT5.mjs";
-import "../../chunk-2IOKI4ES.mjs";
-import "../../chunk-BJTO5JO5.mjs";
+  require_segmentation_models,
+  require_yolo_seg
+} from "../../chunk-ZK7P3TZN.mjs";
+import {
+  require_yolo
+} from "../../chunk-DE7I3VHO.mjs";
+import {
+  require_object_detection_models
+} from "../../chunk-MGT6RUVX.mjs";
+import {
+  require_image_utils
+} from "../../chunk-WG66JYYW.mjs";
+import {
+  require_engine_resolver
+} from "../../chunk-PIFS7AIT.mjs";
+import {
+  __toESM
+} from "../../chunk-3IIFBJCD.mjs";
+
+// src/addons/object-detection/index.ts
+var import_object_detection_models = __toESM(require_object_detection_models());
+var import_segmentation_models = __toESM(require_segmentation_models());
+var import_image_utils = __toESM(require_image_utils());
+var import_yolo = __toESM(require_yolo());
+var import_yolo_seg = __toESM(require_yolo_seg());
+var import_engine_resolver = __toESM(require_engine_resolver());
+import { COCO_TO_MACRO, MACRO_LABELS, COCO_80_LABELS } from "@camstack/types";
+function isSegModel(modelId) {
+  return modelId.includes("-seg");
+}
+var ALL_DETECTION_MODELS = [
+  ...import_object_detection_models.OBJECT_DETECTION_MODELS,
+  ...import_segmentation_models.SEGMENTATION_MODELS
+];
+function applyClassMap(detections, classMap) {
+  return detections.filter((d) => classMap.mapping[d.class] !== void 0).map((d) => ({
+    ...d,
+    originalClass: d.class,
+    class: classMap.mapping[d.class]
+  }));
+}
+var RAM_ESTIMATES = {
+  "yolov8n": 80,
+  "yolov8s": 150,
+  "yolov8s-relu": 150,
+  "yolov8m": 300,
+  "yolov8l": 500,
+  "yolov8x": 800,
+  "yolov9t": 60,
+  "yolov9s": 120,
+  "yolov9c": 300,
+  "yolo11n": 70,
+  "yolo11s": 130,
+  "yolo11m": 280,
+  "yolo11l": 450,
+  "yolo11x": 750,
+  "yolo11n-seg": 84,
+  "yolo11s-seg": 156,
+  "yolo11m-seg": 336,
+  "yolov8n-seg": 96,
+  "yolov8s-seg": 180,
+  "yolov8m-seg": 360
+};
+var ACCURACY_SCORES = {
+  "yolov8n": 55,
+  "yolov8s": 70,
+  "yolov8s-relu": 68,
+  "yolov8m": 82,
+  "yolov8l": 88,
+  "yolov8x": 92,
+  "yolov9t": 58,
+  "yolov9s": 73,
+  "yolov9c": 86,
+  "yolo11n": 62,
+  "yolo11s": 78,
+  "yolo11m": 88,
+  "yolo11l": 93,
+  "yolo11x": 97,
+  "yolo11n-seg": 62,
+  "yolo11s-seg": 78,
+  "yolo11m-seg": 88,
+  "yolov8n-seg": 55,
+  "yolov8s-seg": 70,
+  "yolov8m-seg": 82
+};
+var ObjectDetectionAddon = class {
+  id = "object-detection";
+  slot = "detector";
+  inputClasses = null;
+  outputClasses = ["person", "vehicle", "animal"];
+  slotPriority = 0;
+  manifest = {
+    id: "object-detection",
+    name: "Object Detection",
+    version: "0.1.0",
+    description: "YOLO-based object detection \u2014 detects persons, vehicles, and animals",
+    slot: "detector",
+    inputClasses: void 0,
+    outputClasses: ["person", "vehicle", "animal"],
+    supportsCustomModels: true,
+    mayRequirePython: false,
+    defaultConfig: {
+      modelId: "yolo11n",
+      runtime: "node",
+      backend: "cpu",
+      confidence: 0.5,
+      iouThreshold: 0.45,
+      classMapMode: "macro"
+    }
+  };
+  engine = null;
+  modelEntry;
+  confidence = 0.5;
+  iouThreshold = 0.45;
+  classMapMode = "macro";
+  resolvedConfig = null;
+  ctx = null;
+  getModelRequirements() {
+    return ALL_DETECTION_MODELS.map((m) => ({
+      modelId: m.id,
+      name: m.name,
+      minRAM_MB: RAM_ESTIMATES[m.id] ?? 100,
+      accuracyScore: ACCURACY_SCORES[m.id] ?? 60,
+      formats: Object.keys(m.formats)
+    }));
+  }
+  configure(config) {
+    this.resolvedConfig = config;
+  }
+  async initialize(ctx) {
+    this.ctx = ctx;
+    const cfg = ctx.addonConfig;
+    const modelId = cfg["modelId"] ?? this.resolvedConfig?.modelId ?? "yolo11n";
+    this.confidence = cfg["confidence"] ?? 0.5;
+    this.iouThreshold = cfg["iouThreshold"] ?? 0.45;
+    this.classMapMode = cfg["classMapMode"] ?? "macro";
+    const entry = ALL_DETECTION_MODELS.find((m) => m.id === modelId);
+    if (!entry) {
+      throw new Error(`ObjectDetectionAddon: unknown modelId "${modelId}"`);
+    }
+    this.modelEntry = entry;
+  }
+  async detect(frame) {
+    if (!this.engine) await this.ensureEngine();
+    const start = Date.now();
+    if ("runJpeg" in this.engine && typeof this.engine.runJpeg === "function") {
+      const result = await this.engine.runJpeg(frame.data);
+      const rawDets = result.detections ?? [];
+      const detections2 = rawDets.map((d) => ({
+        class: this.classMapMode === "all" ? d.className : COCO_TO_MACRO.mapping[d.className] ?? d.className,
+        originalClass: d.className,
+        score: d.score,
+        bbox: {
+          x: d.bbox[0] * frame.width,
+          y: d.bbox[1] * frame.height,
+          w: (d.bbox[2] - d.bbox[0]) * frame.width,
+          h: (d.bbox[3] - d.bbox[1]) * frame.height
+        }
+      })).filter((d) => this.classMapMode === "all" || COCO_TO_MACRO.mapping[d.originalClass] !== void 0);
+      return {
+        detections: detections2,
+        inferenceMs: result.inferenceMs ?? Date.now() - start,
+        modelId: this.modelEntry.id
+      };
+    }
+    const { width: inputW, height: inputH } = this.modelEntry.inputSize;
+    const targetSize = Math.max(inputW, inputH);
+    const lb = await (0, import_image_utils.letterbox)(frame.data, targetSize);
+    const numClasses = this.modelEntry.labels.length;
+    const labels = this.modelEntry.labels.map((l) => l.id);
+    const postprocessOpts = {
+      confidence: this.confidence,
+      iouThreshold: this.iouThreshold,
+      labels,
+      scale: lb.scale,
+      padX: lb.padX,
+      padY: lb.padY,
+      originalWidth: lb.originalWidth,
+      originalHeight: lb.originalHeight
+    };
+    let rawDetections;
+    if (isSegModel(this.modelEntry.id)) {
+      const outputs = await this.engine.runMultiOutput(lb.data, [1, 3, targetSize, targetSize]);
+      const outputNames = Object.keys(outputs);
+      if (outputNames.length < 2) {
+        throw new Error(
+          `ObjectDetectionAddon: seg model "${this.modelEntry.id}" returned ${outputNames.length} output(s); expected 2`
+        );
+      }
+      const detectionOutput = outputs[outputNames[0]];
+      const protoOutput = outputs[outputNames[1]];
+      const numMaskCoeffs = 32;
+      const numBoxes = detectionOutput.length / (4 + numClasses + numMaskCoeffs);
+      const maskHeight = 160;
+      const maskWidth = 160;
+      rawDetections = (0, import_yolo_seg.yoloSegPostprocess)(
+        {
+          detectionOutput,
+          protoOutput,
+          numClasses,
+          numBoxes,
+          numMaskCoeffs,
+          maskHeight,
+          maskWidth
+        },
+        postprocessOpts
+      );
+    } else {
+      const output = await this.engine.run(lb.data, [1, 3, targetSize, targetSize]);
+      const numBoxes = output.length / (4 + numClasses);
+      rawDetections = (0, import_yolo.yoloPostprocess)(output, numClasses, numBoxes, postprocessOpts);
+    }
+    const detections = this.classMapMode === "all" ? rawDetections : applyClassMap(rawDetections, COCO_TO_MACRO);
+    return {
+      detections,
+      inferenceMs: Date.now() - start,
+      modelId: this.modelEntry.id
+    };
+  }
+  async ensureEngine() {
+    const config = this.resolvedConfig;
+    const modelId = config?.modelId ?? this.modelEntry.id;
+    const runtime = config?.runtime === "python" ? "coreml" : config?.runtime === "node" ? "onnx" : "auto";
+    const backend = config?.backend ?? "cpu";
+    const format = config?.format ?? "onnx";
+    const entry = ALL_DETECTION_MODELS.find((m) => m.id === modelId) ?? this.modelEntry;
+    this.modelEntry = entry;
+    const modelsDir = this.ctx.models?.getModelsDir() ?? this.ctx.locationPaths.models;
+    if (this.ctx.models) {
+      await this.ctx.models.ensure(modelId, format);
+    }
+    let pythonPath;
+    if (config?.runtime === "python") {
+      for (const cmd of ["python3", "python"]) {
+        try {
+          const { execSync } = await import("child_process");
+          execSync(`${cmd} --version`, { timeout: 3e3, stdio: "ignore" });
+          pythonPath = cmd;
+          break;
+        } catch {
+        }
+      }
+    }
+    const resolved = await (0, import_engine_resolver.resolveEngine)({
+      runtime,
+      backend,
+      modelEntry: entry,
+      modelsDir,
+      pythonPath,
+      models: this.ctx.models
+    });
+    this.engine = resolved.engine;
+  }
+  async shutdown() {
+    await this.engine?.dispose();
+  }
+  getConfigSchema() {
+    return {
+      sections: [
+        {
+          id: "model",
+          title: "Model",
+          columns: 1,
+          fields: [
+            {
+              key: "modelId",
+              label: "Model",
+              type: "model-selector",
+              catalog: [...ALL_DETECTION_MODELS],
+              allowCustom: true,
+              allowConversion: true,
+              acceptFormats: ["onnx", "coreml", "openvino", "tflite"],
+              requiredMetadata: ["inputSize", "labels", "outputFormat"],
+              outputFormatHint: "yolo"
+            }
+          ]
+        },
+        {
+          id: "runtime",
+          title: "Runtime",
+          columns: 2,
+          fields: [
+            {
+              key: "runtime",
+              label: "Runtime",
+              type: "select",
+              options: [
+                { value: "auto", label: "Auto" },
+                { value: "onnx", label: "ONNX Runtime" },
+                { value: "coreml", label: "CoreML (Apple)" },
+                { value: "openvino", label: "OpenVINO (Intel)" }
+              ]
+            },
+            {
+              key: "backend",
+              label: "Backend",
+              type: "select",
+              showWhen: { field: "runtime", equals: "onnx" },
+              options: [
+                { value: "auto", label: "Auto" },
+                { value: "cpu", label: "CPU" },
+                { value: "coreml", label: "CoreML" },
+                { value: "cuda", label: "CUDA (NVIDIA)" },
+                { value: "tensorrt", label: "TensorRT (NVIDIA)" }
+              ]
+            }
+          ]
+        },
+        {
+          id: "thresholds",
+          title: "Detection Thresholds",
+          columns: 2,
+          fields: [
+            {
+              key: "confidence",
+              label: "Confidence Threshold",
+              type: "slider",
+              min: 0.1,
+              max: 1,
+              step: 0.05,
+              default: 0.5
+            },
+            {
+              key: "iouThreshold",
+              label: "IoU Threshold (NMS)",
+              type: "slider",
+              min: 0.1,
+              max: 1,
+              step: 0.05,
+              default: 0.45
+            }
+          ]
+        },
+        {
+          id: "classmap",
+          title: "Class Mapping",
+          columns: 1,
+          fields: [
+            {
+              key: "classMapMode",
+              label: "Output classes",
+              type: "select",
+              options: [
+                { value: "macro", label: "Macro (person / vehicle / animal)" },
+                { value: "all", label: "All COCO classes (80)" }
+              ]
+            }
+          ]
+        }
+      ]
+    };
+  }
+  getClassMap() {
+    return COCO_TO_MACRO;
+  }
+  getModelCatalog() {
+    return [...ALL_DETECTION_MODELS];
+  }
+  getAvailableModels() {
+    return [];
+  }
+  getActiveLabels() {
+    return this.classMapMode === "all" ? COCO_80_LABELS : MACRO_LABELS;
+  }
+  async probe() {
+    return {
+      available: true,
+      runtime: this.engine?.runtime ?? "onnx",
+      device: this.engine?.device ?? "cpu",
+      capabilities: ["fp32"]
+    };
+  }
+};
 export {
   ObjectDetectionAddon as default
 };

package/dist/addons/object-detection/index.mjs.map

@@ -1 +1 @@
-{"version":3,"sources":[],"sourcesContent":[],"mappings":"","names":[]}
+{"version":3,"sources":["../../../src/addons/object-detection/index.ts"],"sourcesContent":["import type {\n  IDetectorProvider,\n  IDetectionAddon,\n  AddonManifest,\n  AddonContext,\n  FrameInput,\n  DetectorOutput,\n  SpatialDetection,\n  ConfigUISchema,\n  ClassMapDefinition,\n  ProbeResult,\n  ModelCatalogEntry,\n  DetectionModel,\n  LabelDefinition,\n  IInferenceEngine,\n  ModelRequirement,\n  ResolvedInferenceConfig,\n} from '@camstack/types'\nimport { OBJECT_DETECTION_MODELS } from '../../catalogs/object-detection-models.js'\nimport { SEGMENTATION_MODELS } from '../../catalogs/segmentation-models.js'\nimport { COCO_TO_MACRO, MACRO_LABELS, COCO_80_LABELS } from '@camstack/types'\nimport { letterbox } from '../../shared/image-utils.js'\nimport { yoloPostprocess } from '../../shared/postprocess/yolo.js'\nimport { yoloSegPostprocess } from '../../shared/postprocess/yolo-seg.js'\nimport { resolveEngine } from '../../shared/engine-resolver.js'\n\n/** Returns true when the model id identifies a YOLO segmentation model (e.g. yolov8n-seg, yolo11m-seg) */\nfunction isSegModel(modelId: string): boolean {\n  return modelId.includes('-seg')\n}\n\n/** Combined catalog: regular detection models + segmentation models */\nconst ALL_DETECTION_MODELS: readonly ModelCatalogEntry[] = [\n  ...OBJECT_DETECTION_MODELS,\n  ...SEGMENTATION_MODELS,\n]\n\n/**\n * classMapMode controls how raw COCO labels are mapped after inference:\n *\n *  - 'macro' (default): apply COCO_TO_MACRO — only person / vehicle / animal\n *    survive, all other classes are dropped.\n *  - 'all': keep every raw COCO label (80 classes). originalClass === class.\n */\ntype ClassMapMode = 'macro' | 'all'\n\n/** Map COCO-class detections to macro categories (person/vehicle/animal) */\nfunction applyClassMap(\n  detections: SpatialDetection[],\n  classMap: ClassMapDefinition,\n): SpatialDetection[] {\n  return detections\n    .filter((d) => classMap.mapping[d.class] !== undefined)\n    .map((d) => ({\n      ...d,\n      originalClass: d.class,\n      class: classMap.mapping[d.class]!,\n    }))\n}\n\n/** RAM estimates per model (MB) */\nconst RAM_ESTIMATES: Record<string, number> = {\n  'yolov8n': 80, 'yolov8s': 150, 'yolov8s-relu': 150, 'yolov8m': 300, 'yolov8l': 500, 'yolov8x': 800,\n  'yolov9t': 60, 'yolov9s': 120, 'yolov9c': 300,\n  'yolo11n': 70, 'yolo11s': 130, 'yolo11m': 280, 'yolo11l': 450, 'yolo11x': 750,\n  'yolo11n-seg': 84, 'yolo11s-seg': 156, 'yolo11m-seg': 336,\n  'yolov8n-seg': 96, 'yolov8s-seg': 180, 'yolov8m-seg': 360,\n}\n\n/**\n * Accuracy scores per model (0-100).\n * YOLO11 scores are higher than YOLOv8 equivalents to reflect the\n * ~2 mAP improvement at same parameter count (e.g. YOLO11n 39.5 vs YOLOv8n 37.3).\n * This ensures the resolver prefers YOLO11 when both families are available.\n */\nconst ACCURACY_SCORES: Record<string, number> = {\n  'yolov8n': 55, 'yolov8s': 70, 'yolov8s-relu': 68, 'yolov8m': 82, 'yolov8l': 88, 'yolov8x': 92,\n  'yolov9t': 58, 'yolov9s': 73, 'yolov9c': 86,\n  'yolo11n': 62, 'yolo11s': 78, 'yolo11m': 88, 'yolo11l': 93, 'yolo11x': 97,\n  'yolo11n-seg': 62, 'yolo11s-seg': 78, 'yolo11m-seg': 88,\n  'yolov8n-seg': 55, 'yolov8s-seg': 70, 'yolov8m-seg': 82,\n}\n\nexport default class ObjectDetectionAddon implements IDetectorProvider, IDetectionAddon {\n  readonly id = 'object-detection'\n  readonly slot = 'detector' as const\n  readonly inputClasses: readonly string[] | null = null\n  readonly outputClasses = ['person', 'vehicle', 'animal'] as const\n  readonly slotPriority = 0\n  readonly manifest: AddonManifest = {\n    id: 'object-detection',\n    name: 'Object Detection',\n    version: '0.1.0',\n\n    description: 'YOLO-based object detection — detects persons, vehicles, and animals',\n\n    slot: 'detector',\n    inputClasses: undefined,\n    outputClasses: ['person', 'vehicle', 'animal'],\n    supportsCustomModels: true,\n    mayRequirePython: false,\n    defaultConfig: {\n      modelId: 'yolo11n',\n      runtime: 'node',\n      backend: 'cpu',\n      confidence: 0.5,\n      iouThreshold: 0.45,\n      classMapMode: 'macro',\n    },\n  }\n\n  private engine: IInferenceEngine | null = null\n  private modelEntry!: ModelCatalogEntry\n  private confidence = 0.5\n  private iouThreshold = 0.45\n  private classMapMode: ClassMapMode = 'macro'\n  private resolvedConfig: ResolvedInferenceConfig | null = null\n  private ctx: AddonContext | null = null\n\n  getModelRequirements(): ModelRequirement[] {\n    return ALL_DETECTION_MODELS.map((m) => ({\n      modelId: m.id,\n      name: m.name,\n      minRAM_MB: RAM_ESTIMATES[m.id] ?? 100,\n      accuracyScore: ACCURACY_SCORES[m.id] ?? 60,\n      formats: Object.keys(m.formats) as readonly string[],\n    }))\n  }\n\n  configure(config: ResolvedInferenceConfig): void {\n    this.resolvedConfig = config\n  }\n\n  async initialize(ctx: AddonContext): Promise<void> {\n    this.ctx = ctx\n    const cfg = ctx.addonConfig\n    const modelId = (cfg['modelId'] as string | undefined) ?? this.resolvedConfig?.modelId ?? 'yolo11n'\n    this.confidence = (cfg['confidence'] as number | undefined) ?? 0.5\n    this.iouThreshold = (cfg['iouThreshold'] as number | undefined) ?? 0.45\n    this.classMapMode = (cfg['classMapMode'] as ClassMapMode | undefined) ?? 'macro'\n\n    const entry = ALL_DETECTION_MODELS.find((m) => m.id === modelId)\n    if (!entry) {\n      throw new Error(`ObjectDetectionAddon: unknown modelId \"${modelId}\"`)\n    }\n    this.modelEntry = entry\n  }\n\n  async detect(frame: FrameInput): Promise<DetectorOutput> {\n    if (!this.engine) await this.ensureEngine()\n    const start = Date.now()\n\n    // Python engine path — send JPEG directly, Python handles preprocess + inference + postprocess\n    if ('runJpeg' in this.engine! && typeof (this.engine as any).runJpeg === 'function') {\n      const result = await (this.engine as any).runJpeg(frame.data) as Record<string, unknown>\n      const rawDets = (result.detections ?? []) as Array<{ className: string; score: number; bbox: number[] }>\n      const detections: SpatialDetection[] = rawDets.map((d) => ({\n        class: this.classMapMode === 'all' ? d.className : (COCO_TO_MACRO.mapping[d.className] ?? d.className),\n        originalClass: d.className,\n        score: d.score,\n        bbox: {\n          x: d.bbox[0]! * frame.width,\n          y: d.bbox[1]! * frame.height,\n          w: (d.bbox[2]! - d.bbox[0]!) * frame.width,\n          h: (d.bbox[3]! - d.bbox[1]!) * frame.height,\n        },\n      })).filter((d) => this.classMapMode === 'all' || COCO_TO_MACRO.mapping[d.originalClass!] !== undefined)\n\n      return {\n        detections,\n        inferenceMs: (result.inferenceMs as number) ?? (Date.now() - start),\n        modelId: this.modelEntry.id,\n      }\n    }\n\n    // Node.js engine path — letterbox + run tensor + postprocess\n    const { width: inputW, height: inputH } = this.modelEntry.inputSize\n    const targetSize = Math.max(inputW, inputH)\n\n    const lb = await letterbox(frame.data, targetSize)\n\n    const numClasses = this.modelEntry.labels.length\n    const labels = this.modelEntry.labels.map((l) => l.id)\n    const postprocessOpts = {\n      confidence: this.confidence,\n      iouThreshold: this.iouThreshold,\n      labels,\n      scale: lb.scale,\n      padX: lb.padX,\n      padY: lb.padY,\n      originalWidth: lb.originalWidth,\n      originalHeight: lb.originalHeight,\n    }\n\n    let rawDetections: SpatialDetection[]\n\n    if (isSegModel(this.modelEntry.id)) {\n      // YOLO-seg models produce two outputs:\n      //   output0: [1, 4 + numClasses + 32, numBoxes]  (detection + mask coefficients)\n      //   output1: [1, 32, 160, 160]                   (prototype masks)\n      const outputs = await this.engine!.runMultiOutput(lb.data, [1, 3, targetSize, targetSize])\n      const outputNames = Object.keys(outputs)\n\n      if (outputNames.length < 2) {\n        throw new Error(\n          `ObjectDetectionAddon: seg model \"${this.modelEntry.id}\" returned ${outputNames.length} output(s); expected 2`,\n        )\n      }\n\n      // ONNX output order is deterministic; first output is the detection tensor, second is protos\n      const detectionOutput = outputs[outputNames[0]!]!\n      const protoOutput = outputs[outputNames[1]!]!\n\n      // Infer dims from tensor sizes\n      const numMaskCoeffs = 32\n      const numBoxes = detectionOutput.length / (4 + numClasses + numMaskCoeffs)\n      const maskHeight = 160\n      const maskWidth = 160\n\n      rawDetections = yoloSegPostprocess(\n        {\n          detectionOutput,\n          protoOutput,\n          numClasses,\n          numBoxes,\n          numMaskCoeffs,\n          maskHeight,\n          maskWidth,\n        },\n        postprocessOpts,\n      )\n    } else {\n      // Standard YOLO output: [1, 4+numClasses, numBoxes]\n      const output = await this.engine!.run(lb.data, [1, 3, targetSize, targetSize])\n      const numBoxes = output.length / (4 + numClasses)\n      rawDetections = yoloPostprocess(output, numClasses, numBoxes, postprocessOpts)\n    }\n\n    const detections = this.classMapMode === 'all'\n      ? rawDetections\n      : applyClassMap(rawDetections, COCO_TO_MACRO)\n\n    return {\n      detections,\n      inferenceMs: Date.now() - start,\n      modelId: this.modelEntry.id,\n    }\n  }\n\n  private async ensureEngine(): Promise<void> {\n    const config = this.resolvedConfig\n    const modelId = config?.modelId ?? this.modelEntry.id\n    const runtime = config?.runtime === 'python' ? 'coreml' : (config?.runtime === 'node' ? 'onnx' : 'auto')\n    const backend = config?.backend ?? 'cpu'\n    const format = config?.format ?? 'onnx'\n\n    const entry = ALL_DETECTION_MODELS.find((m) => m.id === modelId) ?? this.modelEntry\n    this.modelEntry = entry\n\n    const modelsDir = this.ctx!.models?.getModelsDir() ?? this.ctx!.locationPaths.models\n\n    // Download model if needed\n    if (this.ctx!.models) {\n      await this.ctx!.models.ensure(modelId, format as any)\n    }\n\n    // Discover Python path for Python-based runtimes\n    let pythonPath: string | undefined\n    if (config?.runtime === 'python') {\n      for (const cmd of ['python3', 'python']) {\n        try {\n          const { execSync } = await import('node:child_process')\n          execSync(`${cmd} --version`, { timeout: 3000, stdio: 'ignore' })\n          pythonPath = cmd\n          break\n        } catch { /* not found */ }\n      }\n    }\n\n    const resolved = await resolveEngine({\n      runtime: runtime as 'auto',\n      backend,\n      modelEntry: entry,\n      modelsDir,\n      pythonPath,\n      models: this.ctx!.models,\n    })\n    this.engine = resolved.engine\n  }\n\n  async shutdown(): Promise<void> {\n    await this.engine?.dispose()\n  }\n\n  getConfigSchema(): ConfigUISchema {\n    return {\n      sections: [\n        {\n          id: 'model',\n          title: 'Model',\n          columns: 1,\n          fields: [\n            {\n              key: 'modelId',\n              label: 'Model',\n              type: 'model-selector',\n              catalog: [...ALL_DETECTION_MODELS],\n              allowCustom: true,\n              allowConversion: true,\n              acceptFormats: ['onnx', 'coreml', 'openvino', 'tflite'],\n              requiredMetadata: ['inputSize', 'labels', 'outputFormat'],\n              outputFormatHint: 'yolo',\n            },\n          ],\n        },\n        {\n          id: 'runtime',\n          title: 'Runtime',\n          columns: 2,\n          fields: [\n            {\n              key: 'runtime',\n              label: 'Runtime',\n              type: 'select',\n              options: [\n                { value: 'auto', label: 'Auto' },\n                { value: 'onnx', label: 'ONNX Runtime' },\n                { value: 'coreml', label: 'CoreML (Apple)' },\n                { value: 'openvino', label: 'OpenVINO (Intel)' },\n              ],\n            },\n            {\n              key: 'backend',\n              label: 'Backend',\n              type: 'select',\n              showWhen: { field: 'runtime', equals: 'onnx' },\n              options: [\n                { value: 'auto', label: 'Auto' },\n                { value: 'cpu', label: 'CPU' },\n                { value: 'coreml', label: 'CoreML' },\n                { value: 'cuda', label: 'CUDA (NVIDIA)' },\n                { value: 'tensorrt', label: 'TensorRT (NVIDIA)' },\n              ],\n            },\n          ],\n        },\n        {\n          id: 'thresholds',\n          title: 'Detection Thresholds',\n          columns: 2,\n          fields: [\n            {\n              key: 'confidence',\n              label: 'Confidence Threshold',\n              type: 'slider',\n              min: 0.1,\n              max: 1.0,\n              step: 0.05,\n              default: 0.5,\n            },\n            {\n              key: 'iouThreshold',\n              label: 'IoU Threshold (NMS)',\n              type: 'slider',\n              min: 0.1,\n              max: 1.0,\n              step: 0.05,\n              default: 0.45,\n            },\n          ],\n        },\n        {\n          id: 'classmap',\n          title: 'Class Mapping',\n          columns: 1,\n          fields: [\n            {\n              key: 'classMapMode',\n              label: 'Output classes',\n              type: 'select',\n              options: [\n                { value: 'macro', label: 'Macro (person / vehicle / animal)' },\n                { value: 'all', label: 'All COCO classes (80)' },\n              ],\n            },\n          ],\n        },\n      ],\n    }\n  }\n\n  getClassMap(): ClassMapDefinition {\n    return COCO_TO_MACRO\n  }\n\n  getModelCatalog(): ModelCatalogEntry[] {\n    return [...ALL_DETECTION_MODELS]\n  }\n\n  getAvailableModels(): DetectionModel[] {\n    // TODO: check downloaded models in modelsDir\n    return []\n  }\n\n  getActiveLabels(): readonly LabelDefinition[] {\n    return this.classMapMode === 'all' ? COCO_80_LABELS : MACRO_LABELS\n  }\n\n  async probe(): Promise<ProbeResult> {\n    return {\n      available: true,\n      runtime: this.engine?.runtime ?? 'onnx',\n      device: this.engine?.device ?? 'cpu',\n      capabilities: ['fp32'],\n    }\n  }\n}\n"],"mappings":";;;;;;;;;;;;;;;;;;;;;AAkBA,qCAAwC;AACxC,iCAAoC;AAEpC,yBAA0B;AAC1B,kBAAgC;AAChC,sBAAmC;AACnC,6BAA8B;AAJ9B,SAAS,eAAe,cAAc,sBAAsB;AAO5D,SAAS,WAAW,SAA0B;AAC5C,SAAO,QAAQ,SAAS,MAAM;AAChC;AAGA,IAAM,uBAAqD;AAAA,EACzD,GAAG;AAAA,EACH,GAAG;AACL;AAYA,SAAS,cACP,YACA,UACoB;AACpB,SAAO,WACJ,OAAO,CAAC,MAAM,SAAS,QAAQ,EAAE,KAAK,MAAM,MAAS,EACrD,IAAI,CAAC,OAAO;AAAA,IACX,GAAG;AAAA,IACH,eAAe,EAAE;AAAA,IACjB,OAAO,SAAS,QAAQ,EAAE,KAAK;AAAA,EACjC,EAAE;AACN;AAGA,IAAM,gBAAwC;AAAA,EAC5C,WAAW;AAAA,EAAI,WAAW;AAAA,EAAK,gBAAgB;AAAA,EAAK,WAAW;AAAA,EAAK,WAAW;AAAA,EAAK,WAAW;AAAA,EAC/F,WAAW;AAAA,EAAI,WAAW;AAAA,EAAK,WAAW;AAAA,EAC1C,WAAW;AAAA,EAAI,WAAW;AAAA,EAAK,WAAW;AAAA,EAAK,WAAW;AAAA,EAAK,WAAW;AAAA,EAC1E,eAAe;AAAA,EAAI,eAAe;AAAA,EAAK,eAAe;AAAA,EACtD,eAAe;AAAA,EAAI,eAAe;AAAA,EAAK,eAAe;AACxD;AAQA,IAAM,kBAA0C;AAAA,EAC9C,WAAW;AAAA,EAAI,WAAW;AAAA,EAAI,gBAAgB;AAAA,EAAI,WAAW;AAAA,EAAI,WAAW;AAAA,EAAI,WAAW;AAAA,EAC3F,WAAW;AAAA,EAAI,WAAW;AAAA,EAAI,WAAW;AAAA,EACzC,WAAW;AAAA,EAAI,WAAW;AAAA,EAAI,WAAW;AAAA,EAAI,WAAW;AAAA,EAAI,WAAW;AAAA,EACvE,eAAe;AAAA,EAAI,eAAe;AAAA,EAAI,eAAe;AAAA,EACrD,eAAe;AAAA,EAAI,eAAe;AAAA,EAAI,eAAe;AACvD;AAEA,IAAqB,uBAArB,MAAwF;AAAA,EAC7E,KAAK;AAAA,EACL,OAAO;AAAA,EACP,eAAyC;AAAA,EACzC,gBAAgB,CAAC,UAAU,WAAW,QAAQ;AAAA,EAC9C,eAAe;AAAA,EACf,WAA0B;AAAA,IACjC,IAAI;AAAA,IACJ,MAAM;AAAA,IACN,SAAS;AAAA,IAET,aAAa;AAAA,IAEb,MAAM;AAAA,IACN,cAAc;AAAA,IACd,eAAe,CAAC,UAAU,WAAW,QAAQ;AAAA,IAC7C,sBAAsB;AAAA,IACtB,kBAAkB;AAAA,IAClB,eAAe;AAAA,MACb,SAAS;AAAA,MACT,SAAS;AAAA,MACT,SAAS;AAAA,MACT,YAAY;AAAA,MACZ,cAAc;AAAA,MACd,cAAc;AAAA,IAChB;AAAA,EACF;AAAA,EAEQ,SAAkC;AAAA,EAClC;AAAA,EACA,aAAa;AAAA,EACb,eAAe;AAAA,EACf,eAA6B;AAAA,EAC7B,iBAAiD;AAAA,EACjD,MAA2B;AAAA,EAEnC,uBAA2C;AACzC,WAAO,qBAAqB,IAAI,CAAC,OAAO;AAAA,MACtC,SAAS,EAAE;AAAA,MACX,MAAM,EAAE;AAAA,MACR,WAAW,cAAc,EAAE,EAAE,KAAK;AAAA,MAClC,eAAe,gBAAgB,EAAE,EAAE,KAAK;AAAA,MACxC,SAAS,OAAO,KAAK,EAAE,OAAO;AAAA,IAChC,EAAE;AAAA,EACJ;AAAA,EAEA,UAAU,QAAuC;AAC/C,SAAK,iBAAiB;AAAA,EACxB;AAAA,EAEA,MAAM,WAAW,KAAkC;AACjD,SAAK,MAAM;AACX,UAAM,MAAM,IAAI;AAChB,UAAM,UAAW,IAAI,SAAS,KAA4B,KAAK,gBAAgB,WAAW;AAC1F,SAAK,aAAc,IAAI,YAAY,KAA4B;AAC/D,SAAK,eAAgB,IAAI,cAAc,KAA4B;AACnE,SAAK,eAAgB,IAAI,cAAc,KAAkC;AAEzE,UAAM,QAAQ,qBAAqB,KAAK,CAAC,MAAM,EAAE,OAAO,OAAO;AAC/D,QAAI,CAAC,OAAO;AACV,YAAM,IAAI,MAAM,0CAA0C,OAAO,GAAG;AAAA,IACtE;AACA,SAAK,aAAa;AAAA,EACpB;AAAA,EAEA,MAAM,OAAO,OAA4C;AACvD,QAAI,CAAC,KAAK,OAAQ,OAAM,KAAK,aAAa;AAC1C,UAAM,QAAQ,KAAK,IAAI;AAGvB,QAAI,aAAa,KAAK,UAAW,OAAQ,KAAK,OAAe,YAAY,YAAY;AACnF,YAAM,SAAS,MAAO,KAAK,OAAe,QAAQ,MAAM,IAAI;AAC5D,YAAM,UAAW,OAAO,cAAc,CAAC;AACvC,YAAMA,cAAiC,QAAQ,IAAI,CAAC,OAAO;AAAA,QACzD,OAAO,KAAK,iBAAiB,QAAQ,EAAE,YAAa,cAAc,QAAQ,EAAE,SAAS,KAAK,EAAE;AAAA,QAC5F,eAAe,EAAE;AAAA,QACjB,OAAO,EAAE;AAAA,QACT,MAAM;AAAA,UACJ,GAAG,EAAE,KAAK,CAAC,IAAK,MAAM;AAAA,UACtB,GAAG,EAAE,KAAK,CAAC,IAAK,MAAM;AAAA,UACtB,IAAI,EAAE,KAAK,CAAC,IAAK,EAAE,KAAK,CAAC,KAAM,MAAM;AAAA,UACrC,IAAI,EAAE,KAAK,CAAC,IAAK,EAAE,KAAK,CAAC,KAAM,MAAM;AAAA,QACvC;AAAA,MACF,EAAE,EAAE,OAAO,CAAC,MAAM,KAAK,iBAAiB,SAAS,cAAc,QAAQ,EAAE,aAAc,MAAM,MAAS;AAEtG,aAAO;AAAA,QACL,YAAAA;AAAA,QACA,aAAc,OAAO,eAA2B,KAAK,IAAI,IAAI;AAAA,QAC7D,SAAS,KAAK,WAAW;AAAA,MAC3B;AAAA,IACF;AAGA,UAAM,EAAE,OAAO,QAAQ,QAAQ,OAAO,IAAI,KAAK,WAAW;AAC1D,UAAM,aAAa,KAAK,IAAI,QAAQ,MAAM;AAE1C,UAAM,KAAK,UAAM,8BAAU,MAAM,MAAM,UAAU;AAEjD,UAAM,aAAa,KAAK,WAAW,OAAO;AAC1C,UAAM,SAAS,KAAK,WAAW,OAAO,IAAI,CAAC,MAAM,EAAE,EAAE;AACrD,UAAM,kBAAkB;AAAA,MACtB,YAAY,KAAK;AAAA,MACjB,cAAc,KAAK;AAAA,MACnB;AAAA,MACA,OAAO,GAAG;AAAA,MACV,MAAM,GAAG;AAAA,MACT,MAAM,GAAG;AAAA,MACT,eAAe,GAAG;AAAA,MAClB,gBAAgB,GAAG;AAAA,IACrB;AAEA,QAAI;AAEJ,QAAI,WAAW,KAAK,WAAW,EAAE,GAAG;AAIlC,YAAM,UAAU,MAAM,KAAK,OAAQ,eAAe,GAAG,MAAM,CAAC,GAAG,GAAG,YAAY,UAAU,CAAC;AACzF,YAAM,cAAc,OAAO,KAAK,OAAO;AAEvC,UAAI,YAAY,SAAS,GAAG;AAC1B,cAAM,IAAI;AAAA,UACR,oCAAoC,KAAK,WAAW,EAAE,cAAc,YAAY,MAAM;AAAA,QACxF;AAAA,MACF;AAGA,YAAM,kBAAkB,QAAQ,YAAY,CAAC,CAAE;AAC/C,YAAM,cAAc,QAAQ,YAAY,CAAC,CAAE;AAG3C,YAAM,gBAAgB;AACtB,YAAM,WAAW,gBAAgB,UAAU,IAAI,aAAa;AAC5D,YAAM,aAAa;AACnB,YAAM,YAAY;AAElB,0BAAgB;AAAA,QACd;AAAA,UACE;AAAA,UACA;AAAA,UACA;AAAA,UACA;AAAA,UACA;AAAA,UACA;AAAA,UACA;AAAA,QACF;AAAA,QACA;AAAA,MACF;AAAA,IACF,OAAO;AAEL,YAAM,SAAS,MAAM,KAAK,OAAQ,IAAI,GAAG,MAAM,CAAC,GAAG,GAAG,YAAY,UAAU,CAAC;AAC7E,YAAM,WAAW,OAAO,UAAU,IAAI;AACtC,0BAAgB,6BAAgB,QAAQ,YAAY,UAAU,eAAe;AAAA,IAC/E;AAEA,UAAM,aAAa,KAAK,iBAAiB,QACrC,gBACA,cAAc,eAAe,aAAa;AAE9C,WAAO;AAAA,MACL;AAAA,MACA,aAAa,KAAK,IAAI,IAAI;AAAA,MAC1B,SAAS,KAAK,WAAW;AAAA,IAC3B;AAAA,EACF;AAAA,EAEA,MAAc,eAA8B;AAC1C,UAAM,SAAS,KAAK;AACpB,UAAM,UAAU,QAAQ,WAAW,KAAK,WAAW;AACnD,UAAM,UAAU,QAAQ,YAAY,WAAW,WAAY,QAAQ,YAAY,SAAS,SAAS;AACjG,UAAM,UAAU,QAAQ,WAAW;AACnC,UAAM,SAAS,QAAQ,UAAU;AAEjC,UAAM,QAAQ,qBAAqB,KAAK,CAAC,MAAM,EAAE,OAAO,OAAO,KAAK,KAAK;AACzE,SAAK,aAAa;AAElB,UAAM,YAAY,KAAK,IAAK,QAAQ,aAAa,KAAK,KAAK,IAAK,cAAc;AAG9E,QAAI,KAAK,IAAK,QAAQ;AACpB,YAAM,KAAK,IAAK,OAAO,OAAO,SAAS,MAAa;AAAA,IACtD;AAGA,QAAI;AACJ,QAAI,QAAQ,YAAY,UAAU;AAChC,iBAAW,OAAO,CAAC,WAAW,QAAQ,GAAG;AACvC,YAAI;AACF,gBAAM,EAAE,SAAS,IAAI,MAAM,OAAO,eAAoB;AACtD,mBAAS,GAAG,GAAG,cAAc,EAAE,SAAS,KAAM,OAAO,SAAS,CAAC;AAC/D,uBAAa;AACb;AAAA,QACF,QAAQ;AAAA,QAAkB;AAAA,MAC5B;AAAA,IACF;AAEA,UAAM,WAAW,UAAM,sCAAc;AAAA,MACnC;AAAA,MACA;AAAA,MACA,YAAY;AAAA,MACZ;AAAA,MACA;AAAA,MACA,QAAQ,KAAK,IAAK;AAAA,IACpB,CAAC;AACD,SAAK,SAAS,SAAS;AAAA,EACzB;AAAA,EAEA,MAAM,WAA0B;AAC9B,UAAM,KAAK,QAAQ,QAAQ;AAAA,EAC7B;AAAA,EAEA,kBAAkC;AAChC,WAAO;AAAA,MACL,UAAU;AAAA,QACR;AAAA,UACE,IAAI;AAAA,UACJ,OAAO;AAAA,UACP,SAAS;AAAA,UACT,QAAQ;AAAA,YACN;AAAA,cACE,KAAK;AAAA,cACL,OAAO;AAAA,cACP,MAAM;AAAA,cACN,SAAS,CAAC,GAAG,oBAAoB;AAAA,cACjC,aAAa;AAAA,cACb,iBAAiB;AAAA,cACjB,eAAe,CAAC,QAAQ,UAAU,YAAY,QAAQ;AAAA,cACtD,kBAAkB,CAAC,aAAa,UAAU,cAAc;AAAA,cACxD,kBAAkB;AAAA,YACpB;AAAA,UACF;AAAA,QACF;AAAA,QACA;AAAA,UACE,IAAI;AAAA,UACJ,OAAO;AAAA,UACP,SAAS;AAAA,UACT,QAAQ;AAAA,YACN;AAAA,cACE,KAAK;AAAA,cACL,OAAO;AAAA,cACP,MAAM;AAAA,cACN,SAAS;AAAA,gBACP,EAAE,OAAO,QAAQ,OAAO,OAAO;AAAA,gBAC/B,EAAE,OAAO,QAAQ,OAAO,eAAe;AAAA,gBACvC,EAAE,OAAO,UAAU,OAAO,iBAAiB;AAAA,gBAC3C,EAAE,OAAO,YAAY,OAAO,mBAAmB;AAAA,cACjD;AAAA,YACF;AAAA,YACA;AAAA,cACE,KAAK;AAAA,cACL,OAAO;AAAA,cACP,MAAM;AAAA,cACN,UAAU,EAAE,OAAO,WAAW,QAAQ,OAAO;AAAA,cAC7C,SAAS;AAAA,gBACP,EAAE,OAAO,QAAQ,OAAO,OAAO;AAAA,gBAC/B,EAAE,OAAO,OAAO,OAAO,MAAM;AAAA,gBAC7B,EAAE,OAAO,UAAU,OAAO,SAAS;AAAA,gBACnC,EAAE,OAAO,QAAQ,OAAO,gBAAgB;AAAA,gBACxC,EAAE,OAAO,YAAY,OAAO,oBAAoB;AAAA,cAClD;AAAA,YACF;AAAA,UACF;AAAA,QACF;AAAA,QACA;AAAA,UACE,IAAI;AAAA,UACJ,OAAO;AAAA,UACP,SAAS;AAAA,UACT,QAAQ;AAAA,YACN;AAAA,cACE,KAAK;AAAA,cACL,OAAO;AAAA,cACP,MAAM;AAAA,cACN,KAAK;AAAA,cACL,KAAK;AAAA,cACL,MAAM;AAAA,cACN,SAAS;AAAA,YACX;AAAA,YACA;AAAA,cACE,KAAK;AAAA,cACL,OAAO;AAAA,cACP,MAAM;AAAA,cACN,KAAK;AAAA,cACL,KAAK;AAAA,cACL,MAAM;AAAA,cACN,SAAS;AAAA,YACX;AAAA,UACF;AAAA,QACF;AAAA,QACA;AAAA,UACE,IAAI;AAAA,UACJ,OAAO;AAAA,UACP,SAAS;AAAA,UACT,QAAQ;AAAA,YACN;AAAA,cACE,KAAK;AAAA,cACL,OAAO;AAAA,cACP,MAAM;AAAA,cACN,SAAS;AAAA,gBACP,EAAE,OAAO,SAAS,OAAO,oCAAoC;AAAA,gBAC7D,EAAE,OAAO,OAAO,OAAO,wBAAwB;AAAA,cACjD;AAAA,YACF;AAAA,UACF;AAAA,QACF;AAAA,MACF;AAAA,IACF;AAAA,EACF;AAAA,EAEA,cAAkC;AAChC,WAAO;AAAA,EACT;AAAA,EAEA,kBAAuC;AACrC,WAAO,CAAC,GAAG,oBAAoB;AAAA,EACjC;AAAA,EAEA,qBAAuC;AAErC,WAAO,CAAC;AAAA,EACV;AAAA,EAEA,kBAA8C;AAC5C,WAAO,KAAK,iBAAiB,QAAQ,iBAAiB;AAAA,EACxD;AAAA,EAEA,MAAM,QAA8B;AAClC,WAAO;AAAA,MACL,WAAW;AAAA,MACX,SAAS,KAAK,QAAQ,WAAW;AAAA,MACjC,QAAQ,KAAK,QAAQ,UAAU;AAAA,MAC/B,cAAc,CAAC,MAAM;AAAA,IACvB;AAAA,EACF;AACF;","names":["detections"]}