@richardmcquiston01/ai-tools 0.0.0-stage → 0.2.0

package/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2026 Richard McQuiston
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

package/README.md

@@ -1,3 +1,79 @@
-# Temporary Holding Version
+# @richardmcquiston01/ai-tools
 
-This version is a temporary placeholder for this package. An operational version to replace this has been submitted for review and is awaiting a staged release.
+In-browser AI image tools for laser, CNC and other maker-machine workflows: background removal, upscaling and subject-aware auto-crop. Everything runs client-side — no API keys and no server.
+
+Part of [maker-image-tools](https://github.com/RichardMcQuiston01/maker-image-tools). Built on [`@richardmcquiston01/core-image`](https://www.npmjs.com/package/@richardmcquiston01/core-image).
+
+## Install
+
+```sh
+npm i @richardmcquiston01/ai-tools
+```
+
+Requires an environment with `ImageData` (browser, or a polyfill in Node). ESM only.
+
+## Usage
+
+```ts
+import {
+  computeCropRegion,
+  loadBackgroundRemovalModel,
+  loadUpscaleModel,
+} from "@richardmcquiston01/ai-tools";
+
+// Background removal + auto-crop (U²-Net-P via onnxruntime-web).
+// Pass a URL, ArrayBuffer or Uint8Array of the bundled `models/u2netp.onnx`.
+const bg = await loadBackgroundRemovalModel("/models/u2netp.onnx", {
+  wasmPaths: "/ort/", // where ort-wasm*.wasm is served from (bundlers like Vite need this)
+});
+const cutout: ImageData = await bg.removeBackground(image);
+
+const mask = await bg.computeSaliencyMask(image);
+const region = computeCropRegion(mask, { threshold: 0.5, margin: 8 });
+
+// Upscaling (ESRGAN-slim via TensorFlow.js).
+const upscaler = await loadUpscaleModel(4);
+const bigger: ImageData = await upscaler.upscale(image);
+```
+
+## As an `ImageGenerator` (optional)
+
+`@richardmcquiston01/ai-tools/provider` wraps background removal as an `ImageGenerator` from [`@richardmcquiston01/ai-providers`](https://www.npmjs.com/package/@richardmcquiston01/ai-providers), so an app can offer it next to other providers (a local ComfyUI, a hosted API) and swap between them through one interface. `ai-providers` is an optional peer dependency: install it only if you use this entry point.
+
+```ts
+import {
+  browserImageCodec,
+  lazyBackgroundRemovalModel,
+  OnnxBackgroundRemover,
+  REMOVE_BACKGROUND_MODE,
+} from "@richardmcquiston01/ai-tools/provider";
+
+const remover = new OnnxBackgroundRemover({
+  // ONNX Runtime is imported on first use, so listing the provider costs nothing.
+  loadModel: lazyBackgroundRemovalModel("/models/u2netp.onnx"),
+  // Providers exchange encoded images; the codec turns them into pixels and back.
+  codec: browserImageCodec,
+});
+
+const result = await remover.generateImage({
+  instructions: "",
+  prompt: "",
+  images: [{ mimeType: "image/png", data: pngBytes }],
+  mode: REMOVE_BACKGROUND_MODE,
+});
+if (result.ok)
+  use(result.value.image); // a PNG with a transparent background
+else console.error(result.error); // names the provider and what failed
+```
+
+`browserImageCodec` needs the DOM (canvas); pass your own `ImageCodec` elsewhere. The model is loaded once and shared between runs, and a failed load is retried on the next call.
+
+## Model files and offline use
+
+- **Background removal:** the package ships `models/u2netp.onnx` (about 4.5 MB). Serve or import it yourself and pass it to `loadBackgroundRemovalModel`; nothing is fetched on your behalf.
+- **Upscaling:** by default the model definitions load their weights from a public CDN mirror of `@upscalerjs/esrgan-slim`. For a self-contained deployment, serve that package's `models/<scale>x/` directory and pass the resolved `model.json` URL as `modelUrl`.
+- **Size:** TensorFlow.js and ONNX Runtime are large. Load this package lazily (`await import(...)`) rather than in your main bundle.
+
+## Licenses
+
+The code is MIT (see `LICENSE`). The bundled `models/u2netp.onnx` is the Apache-2.0 U²-Net-P checkpoint; its attribution is in `models/NOTICE.md` and the licence text is in `models/LICENSE-APACHE-2.0.txt`.

package/dist/auto-crop.d.ts

@@ -0,0 +1,26 @@
+import type { SaliencyMask } from "./background-removal.js";
+export interface CropRegion {
+    x: number;
+    y: number;
+    width: number;
+    height: number;
+}
+export interface ComputeCropRegionOptions {
+    /**
+     * Foreground-probability threshold in [0, 1] above which a pixel counts as
+     * "subject". Defaults to 0.5.
+     */
+    threshold?: number;
+    /**
+     * Extra padding added on each side of the tight subject bounding box, in
+     * pixels. Defaults to 0.
+     */
+    margin?: number;
+}
+/**
+ * Computes the tight (plus optional margin) bounding box around the
+ * foreground of a saliency mask, clamped to the mask's own bounds. Falls
+ * back to the full mask extent if no pixel clears the threshold.
+ */
+export declare function computeCropRegion(mask: SaliencyMask, options?: ComputeCropRegionOptions): CropRegion;
+//# sourceMappingURL=auto-crop.d.ts.map

package/dist/auto-crop.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"auto-crop.d.ts","sourceRoot":"","sources":["../src/auto-crop.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,EAAE,YAAY,EAAE,MAAM,yBAAyB,CAAC;AAE5D,MAAM,WAAW,UAAU;IACzB,CAAC,EAAE,MAAM,CAAC;IACV,CAAC,EAAE,MAAM,CAAC;IACV,KAAK,EAAE,MAAM,CAAC;IACd,MAAM,EAAE,MAAM,CAAC;CAChB;AAED,MAAM,WAAW,wBAAwB;IACvC;;;OAGG;IACH,SAAS,CAAC,EAAE,MAAM,CAAC;IACnB;;;OAGG;IACH,MAAM,CAAC,EAAE,MAAM,CAAC;CACjB;AAED;;;;GAIG;AACH,wBAAgB,iBAAiB,CAC/B,IAAI,EAAE,YAAY,EAClB,OAAO,CAAC,EAAE,wBAAwB,GACjC,UAAU,CAqCZ"}

package/dist/auto-crop.js

@@ -0,0 +1,43 @@
+/**
+ * Computes the tight (plus optional margin) bounding box around the
+ * foreground of a saliency mask, clamped to the mask's own bounds. Falls
+ * back to the full mask extent if no pixel clears the threshold.
+ */
+export function computeCropRegion(mask, options) {
+    const threshold = options?.threshold ?? 0.5;
+    const margin = options?.margin ?? 0;
+    const { width, height, values } = mask;
+    let minX = width;
+    let minY = height;
+    let maxX = -1;
+    let maxY = -1;
+    for (let y = 0; y < height; y++) {
+        for (let x = 0; x < width; x++) {
+            const value = values[y * width + x] ?? 0;
+            if (value >= threshold) {
+                if (x < minX)
+                    minX = x;
+                if (x > maxX)
+                    maxX = x;
+                if (y < minY)
+                    minY = y;
+                if (y > maxY)
+                    maxY = y;
+            }
+        }
+    }
+    if (maxX < minX || maxY < minY) {
+        return { x: 0, y: 0, width, height };
+    }
+    const x = Math.max(0, minX - margin);
+    const y = Math.max(0, minY - margin);
+    const clampedMaxX = Math.min(width - 1, maxX + margin);
+    const clampedMaxY = Math.min(height - 1, maxY + margin);
+    return {
+        x,
+        y,
+        width: clampedMaxX - x + 1,
+        height: clampedMaxY - y + 1,
+    };
+}
+//# sourceMappingURL=auto-crop.js.map

package/dist/auto-crop.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"auto-crop.js","sourceRoot":"","sources":["../src/auto-crop.ts"],"names":[],"mappings":"AAsBA;;;;GAIG;AACH,MAAM,UAAU,iBAAiB,CAC/B,IAAkB,EAClB,OAAkC;IAElC,MAAM,SAAS,GAAG,OAAO,EAAE,SAAS,IAAI,GAAG,CAAC;IAC5C,MAAM,MAAM,GAAG,OAAO,EAAE,MAAM,IAAI,CAAC,CAAC;IACpC,MAAM,EAAE,KAAK,EAAE,MAAM,EAAE,MAAM,EAAE,GAAG,IAAI,CAAC;IAEvC,IAAI,IAAI,GAAG,KAAK,CAAC;IACjB,IAAI,IAAI,GAAG,MAAM,CAAC;IAClB,IAAI,IAAI,GAAG,CAAC,CAAC,CAAC;IACd,IAAI,IAAI,GAAG,CAAC,CAAC,CAAC;IAEd,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,MAAM,EAAE,CAAC,EAAE,EAAE,CAAC;QAChC,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,KAAK,EAAE,CAAC,EAAE,EAAE,CAAC;YAC/B,MAAM,KAAK,GAAG,MAAM,CAAC,CAAC,GAAG,KAAK,GAAG,CAAC,CAAC,IAAI,CAAC,CAAC;YACzC,IAAI,KAAK,IAAI,SAAS,EAAE,CAAC;gBACvB,IAAI,CAAC,GAAG,IAAI;oBAAE,IAAI,GAAG,CAAC,CAAC;gBACvB,IAAI,CAAC,GAAG,IAAI;oBAAE,IAAI,GAAG,CAAC,CAAC;gBACvB,IAAI,CAAC,GAAG,IAAI;oBAAE,IAAI,GAAG,CAAC,CAAC;gBACvB,IAAI,CAAC,GAAG,IAAI;oBAAE,IAAI,GAAG,CAAC,CAAC;YACzB,CAAC;QACH,CAAC;IACH,CAAC;IAED,IAAI,IAAI,GAAG,IAAI,IAAI,IAAI,GAAG,IAAI,EAAE,CAAC;QAC/B,OAAO,EAAE,CAAC,EAAE,CAAC,EAAE,CAAC,EAAE,CAAC,EAAE,KAAK,EAAE,MAAM,EAAE,CAAC;IACvC,CAAC;IAED,MAAM,CAAC,GAAG,IAAI,CAAC,GAAG,CAAC,CAAC,EAAE,IAAI,GAAG,MAAM,CAAC,CAAC;IACrC,MAAM,CAAC,GAAG,IAAI,CAAC,GAAG,CAAC,CAAC,EAAE,IAAI,GAAG,MAAM,CAAC,CAAC;IACrC,MAAM,WAAW,GAAG,IAAI,CAAC,GAAG,CAAC,KAAK,GAAG,CAAC,EAAE,IAAI,GAAG,MAAM,CAAC,CAAC;IACvD,MAAM,WAAW,GAAG,IAAI,CAAC,GAAG,CAAC,MAAM,GAAG,CAAC,EAAE,IAAI,GAAG,MAAM,CAAC,CAAC;IAExD,OAAO;QACL,CAAC;QACD,CAAC;QACD,KAAK,EAAE,WAAW,GAAG,CAAC,GAAG,CAAC;QAC1B,MAAM,EAAE,WAAW,GAAG,CAAC,GAAG,CAAC;KAC5B,CAAC;AACJ,CAAC"}

package/dist/background-removal.d.ts

@@ -0,0 +1,33 @@
+export interface LoadBackgroundRemovalModelOptions {
+    /**
+     * Base URL onnxruntime-web should use to locate its own WASM runtime files
+     * (ort-wasm*.wasm), passed through to `ort.env.wasm.wasmPaths`. Bundlers
+     * that don't automatically resolve onnxruntime-web's WASM assets (e.g.
+     * Vite, unless configured) need this pointed at wherever those files were
+     * copied. If omitted, onnxruntime-web's own default resolution is used.
+     */
+    wasmPaths?: string;
+}
+export interface SaliencyMask {
+    /** Row-major foreground-probability values in [0, 1], one per pixel of the source image. */
+    values: Float32Array;
+    width: number;
+    height: number;
+}
+export interface BackgroundRemovalModel {
+    /**
+     * Removes the background from `image`, returning a same-size ImageData
+     * whose background pixels are made transparent (alpha scaled by the
+     * predicted foreground mask).
+     */
+    removeBackground(image: ImageData): Promise<ImageData>;
+    /**
+     * Runs the same underlying saliency model as `removeBackground` but
+     * returns the raw upsampled mask instead of compositing it into an image —
+     * useful for anything that just needs to know *where* the subject is (e.g.
+     * auto-crop) without paying for a full alpha-composite pass.
+     */
+    computeSaliencyMask(image: ImageData): Promise<SaliencyMask>;
+}
+export declare function loadBackgroundRemovalModel(model: string | ArrayBuffer | Uint8Array, options?: LoadBackgroundRemovalModelOptions): Promise<BackgroundRemovalModel>;
+//# sourceMappingURL=background-removal.d.ts.map

package/dist/background-removal.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"background-removal.d.ts","sourceRoot":"","sources":["../src/background-removal.ts"],"names":[],"mappings":"AAaA,MAAM,WAAW,iCAAiC;IAChD;;;;;;OAMG;IACH,SAAS,CAAC,EAAE,MAAM,CAAC;CACpB;AAED,MAAM,WAAW,YAAY;IAC3B,4FAA4F;IAC5F,MAAM,EAAE,YAAY,CAAC;IACrB,KAAK,EAAE,MAAM,CAAC;IACd,MAAM,EAAE,MAAM,CAAC;CAChB;AAED,MAAM,WAAW,sBAAsB;IACrC;;;;OAIG;IACH,gBAAgB,CAAC,KAAK,EAAE,SAAS,GAAG,OAAO,CAAC,SAAS,CAAC,CAAC;IACvD;;;;;OAKG;IACH,mBAAmB,CAAC,KAAK,EAAE,SAAS,GAAG,OAAO,CAAC,YAAY,CAAC,CAAC;CAC9D;AAoED,wBAAsB,0BAA0B,CAC9C,KAAK,EAAE,MAAM,GAAG,WAAW,GAAG,UAAU,EACxC,OAAO,CAAC,EAAE,iCAAiC,GAC1C,OAAO,CAAC,sBAAsB,CAAC,CAoDjC"}

package/dist/background-removal.js

@@ -0,0 +1,116 @@
+import { createImageData, resize } from "@richardmcquiston01/core-image";
+import * as ort from "onnxruntime-web";
+/**
+ * U2NETP (see models/NOTICE.md) expects a 320x320 RGB input, normalized with
+ * ImageNet mean/std, laid out channel-first (NCHW). Its first output is the
+ * final fused, sigmoid-activated saliency mask; the remaining six outputs are
+ * the network's deep-supervision side outputs and are not used here.
+ */
+const MODEL_INPUT_SIZE = 320;
+const CHANNEL_MEAN = [0.485, 0.456, 0.406];
+const CHANNEL_STD = [0.229, 0.224, 0.225];
+function preprocess(image) {
+    const resized = resize(image, {
+        width: MODEL_INPUT_SIZE,
+        height: MODEL_INPUT_SIZE,
+        method: "bilinear",
+    });
+    const pixelCount = MODEL_INPUT_SIZE * MODEL_INPUT_SIZE;
+    const chw = new Float32Array(3 * pixelCount);
+    const data = resized.data;
+    for (let i = 0; i < pixelCount; i++) {
+        const r = (data[i * 4] ?? 0) / 255;
+        const g = (data[i * 4 + 1] ?? 0) / 255;
+        const b = (data[i * 4 + 2] ?? 0) / 255;
+        chw[i] = (r - CHANNEL_MEAN[0]) / CHANNEL_STD[0];
+        chw[pixelCount + i] = (g - CHANNEL_MEAN[1]) / CHANNEL_STD[1];
+        chw[pixelCount * 2 + i] = (b - CHANNEL_MEAN[2]) / CHANNEL_STD[2];
+    }
+    return new ort.Tensor("float32", chw, [1, 3, MODEL_INPUT_SIZE, MODEL_INPUT_SIZE]);
+}
+/** Min-max normalizes the raw mask output to [0, 1], matching U2Net's standard postprocessing. */
+function normalizeMask(mask) {
+    let min = Infinity;
+    let max = -Infinity;
+    for (const v of mask) {
+        if (v < min)
+            min = v;
+        if (v > max)
+            max = v;
+    }
+    const range = max - min;
+    if (range < 1e-6) {
+        return mask;
+    }
+    const normalized = new Float32Array(mask.length);
+    for (let i = 0; i < mask.length; i++) {
+        normalized[i] = ((mask[i] ?? 0) - min) / range;
+    }
+    return normalized;
+}
+/** Upsamples a 320x320 [0,1] mask to `width`x`height` by round-tripping it through the RGBA resize filter. */
+function upsampleMask(mask, width, height) {
+    const maskImage = createImageData(MODEL_INPUT_SIZE, MODEL_INPUT_SIZE);
+    const maskData = maskImage.data;
+    for (let i = 0; i < mask.length; i++) {
+        const value = Math.round((mask[i] ?? 0) * 255);
+        maskData[i * 4] = value;
+        maskData[i * 4 + 1] = value;
+        maskData[i * 4 + 2] = value;
+        maskData[i * 4 + 3] = 255;
+    }
+    const resized = resize(maskImage, { width, height, method: "bilinear" });
+    const result = new Float32Array(width * height);
+    for (let i = 0; i < result.length; i++) {
+        result[i] = (resized.data[i * 4] ?? 0) / 255;
+    }
+    return result;
+}
+function supportsWebGpu() {
+    return typeof navigator !== "undefined" && "gpu" in navigator && navigator.gpu != null;
+}
+export async function loadBackgroundRemovalModel(model, options) {
+    if (options?.wasmPaths !== undefined) {
+        ort.env.wasm.wasmPaths = options.wasmPaths;
+    }
+    const executionProviders = supportsWebGpu() ? ["webgpu", "wasm"] : ["wasm"];
+    const session = typeof model === "string"
+        ? await ort.InferenceSession.create(model, { executionProviders })
+        : await ort.InferenceSession.create(model instanceof Uint8Array ? model : new Uint8Array(model), { executionProviders });
+    const rawInputName = session.inputNames[0];
+    const rawOutputName = session.outputNames[0];
+    if (!rawInputName || !rawOutputName) {
+        throw new Error("Background removal model has no input/output names");
+    }
+    const inputName = rawInputName;
+    const outputName = rawOutputName;
+    async function computeSaliencyMask(image) {
+        const tensor = preprocess(image);
+        const results = await session.run({ [inputName]: tensor });
+        const output = results[outputName];
+        if (!output) {
+            throw new Error(`Background removal model did not produce output "${outputName}"`);
+        }
+        const mask = normalizeMask(output.data);
+        const values = upsampleMask(mask, image.width, image.height);
+        return { values, width: image.width, height: image.height };
+    }
+    return {
+        computeSaliencyMask,
+        async removeBackground(image) {
+            const { values: upsampled } = await computeSaliencyMask(image);
+            const result = createImageData(image.width, image.height);
+            const srcData = image.data;
+            const dstData = result.data;
+            for (let i = 0; i < upsampled.length; i++) {
+                const srcAlpha = srcData[i * 4 + 3] ?? 255;
+                dstData[i * 4] = srcData[i * 4] ?? 0;
+                dstData[i * 4 + 1] = srcData[i * 4 + 1] ?? 0;
+                dstData[i * 4 + 2] = srcData[i * 4 + 2] ?? 0;
+                dstData[i * 4 + 3] = Math.round((upsampled[i] ?? 0) * srcAlpha);
+            }
+            return result;
+        },
+    };
+}
+//# sourceMappingURL=background-removal.js.map

package/dist/background-removal.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"background-removal.js","sourceRoot":"","sources":["../src/background-removal.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,eAAe,EAAE,MAAM,EAAE,MAAM,gCAAgC,CAAC;AACzE,OAAO,KAAK,GAAG,MAAM,iBAAiB,CAAC;AAEvC;;;;;GAKG;AACH,MAAM,gBAAgB,GAAG,GAAG,CAAC;AAC7B,MAAM,YAAY,GAAG,CAAC,KAAK,EAAE,KAAK,EAAE,KAAK,CAAC,CAAC;AAC3C,MAAM,WAAW,GAAG,CAAC,KAAK,EAAE,KAAK,EAAE,KAAK,CAAC,CAAC;AAoC1C,SAAS,UAAU,CAAC,KAAgB;IAClC,MAAM,OAAO,GAAG,MAAM,CAAC,KAAK,EAAE;QAC5B,KAAK,EAAE,gBAAgB;QACvB,MAAM,EAAE,gBAAgB;QACxB,MAAM,EAAE,UAAU;KACnB,CAAC,CAAC;IAEH,MAAM,UAAU,GAAG,gBAAgB,GAAG,gBAAgB,CAAC;IACvD,MAAM,GAAG,GAAG,IAAI,YAAY,CAAC,CAAC,GAAG,UAAU,CAAC,CAAC;IAC7C,MAAM,IAAI,GAAG,OAAO,CAAC,IAAI,CAAC;IAE1B,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,UAAU,EAAE,CAAC,EAAE,EAAE,CAAC;QACpC,MAAM,CAAC,GAAG,CAAC,IAAI,CAAC,CAAC,GAAG,CAAC,CAAC,IAAI,CAAC,CAAC,GAAG,GAAG,CAAC;QACnC,MAAM,CAAC,GAAG,CAAC,IAAI,CAAC,CAAC,GAAG,CAAC,GAAG,CAAC,CAAC,IAAI,CAAC,CAAC,GAAG,GAAG,CAAC;QACvC,MAAM,CAAC,GAAG,CAAC,IAAI,CAAC,CAAC,GAAG,CAAC,GAAG,CAAC,CAAC,IAAI,CAAC,CAAC,GAAG,GAAG,CAAC;QACvC,GAAG,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,GAAG,YAAY,CAAC,CAAC,CAAE,CAAC,GAAG,WAAW,CAAC,CAAC,CAAE,CAAC;QAClD,GAAG,CAAC,UAAU,GAAG,CAAC,CAAC,GAAG,CAAC,CAAC,GAAG,YAAY,CAAC,CAAC,CAAE,CAAC,GAAG,WAAW,CAAC,CAAC,CAAE,CAAC;QAC/D,GAAG,CAAC,UAAU,GAAG,CAAC,GAAG,CAAC,CAAC,GAAG,CAAC,CAAC,GAAG,YAAY,CAAC,CAAC,CAAE,CAAC,GAAG,WAAW,CAAC,CAAC,CAAE,CAAC;IACrE,CAAC;IAED,OAAO,IAAI,GAAG,CAAC,MAAM,CAAC,SAAS,EAAE,GAAG,EAAE,CAAC,CAAC,EAAE,CAAC,EAAE,gBAAgB,EAAE,gBAAgB,CAAC,CAAC,CAAC;AACpF,CAAC;AAED,kGAAkG;AAClG,SAAS,aAAa,CAAC,IAAkB;IACvC,IAAI,GAAG,GAAG,QAAQ,CAAC;IACnB,IAAI,GAAG,GAAG,CAAC,QAAQ,CAAC;IACpB,KAAK,MAAM,CAAC,IAAI,IAAI,EAAE,CAAC;QACrB,IAAI,CAAC,GAAG,GAAG;YAAE,GAAG,GAAG,CAAC,CAAC;QACrB,IAAI,CAAC,GAAG,GAAG;YAAE,GAAG,GAAG,CAAC,CAAC;IACvB,CAAC;IACD,MAAM,KAAK,GAAG,GAAG,GAAG,GAAG,CAAC;IACxB,IAAI,KAAK,GAAG,IAAI,EAAE,CAAC;QACjB,OAAO,IAAI,CAAC;IACd,CAAC;IACD,MAAM,UAAU,GAAG,IAAI,YAAY,CAAC,IAAI,CAAC,MAAM,CAAC,CAAC;IACjD,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,IAAI,CAAC,MAAM,EAAE,CAAC,EAAE,EAAE,CAAC;QACrC,UAAU,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,IAAI,CAAC,CAAC,GAAG,GAAG,CAAC,GAAG,KAAK,CAAC;IACjD,CAAC;IACD,OAAO,UAAU,CAAC;AACpB,CAAC;AAED,8GAA8G;AAC9G,SAAS,YAAY,CAAC,IAAkB,EAAE,KAAa,EAAE,MAAc;IACrE,MAAM,SAAS,GAAG,eAAe,CAAC,gBAAgB,EAAE,gBAAgB,CAAC,CAAC;IACtE,MAAM,QAAQ,GAAG,SAAS,CAAC,IAAI,CAAC;IAChC,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,IAAI,CAAC,MAAM,EAAE,CAAC,EAAE,EAAE,CAAC;QACrC,MAAM,KAAK,GAAG,IAAI,CAAC,KAAK,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,IAAI,CAAC,CAAC,GAAG,GAAG,CAAC,CAAC;QAC/C,QAAQ,CAAC,CAAC,GAAG,CAAC,CAAC,GAAG,KAAK,CAAC;QACxB,QAAQ,CAAC,CAAC,GAAG,CAAC,GAAG,CAAC,CAAC,GAAG,KAAK,CAAC;QAC5B,QAAQ,CAAC,CAAC,GAAG,CAAC,GAAG,CAAC,CAAC,GAAG,KAAK,CAAC;QAC5B,QAAQ,CAAC,CAAC,GAAG,CAAC,GAAG,CAAC,CAAC,GAAG,GAAG,CAAC;IAC5B,CAAC;IAED,MAAM,OAAO,GAAG,MAAM,CAAC,SAAS,EAAE,EAAE,KAAK,EAAE,MAAM,EAAE,MAAM,EAAE,UAAU,EAAE,CAAC,CAAC;IACzE,MAAM,MAAM,GAAG,IAAI,YAAY,CAAC,KAAK,GAAG,MAAM,CAAC,CAAC;IAChD,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,MAAM,CAAC,MAAM,EAAE,CAAC,EAAE,EAAE,CAAC;QACvC,MAAM,CAAC,CAAC,CAAC,GAAG,CAAC,OAAO,CAAC,IAAI,CAAC,CAAC,GAAG,CAAC,CAAC,IAAI,CAAC,CAAC,GAAG,GAAG,CAAC;IAC/C,CAAC;IACD,OAAO,MAAM,CAAC;AAChB,CAAC;AAED,SAAS,cAAc;IACrB,OAAO,OAAO,SAAS,KAAK,WAAW,IAAI,KAAK,IAAI,SAAS,IAAI,SAAS,CAAC,GAAG,IAAI,IAAI,CAAC;AACzF,CAAC;AAED,MAAM,CAAC,KAAK,UAAU,0BAA0B,CAC9C,KAAwC,EACxC,OAA2C;IAE3C,IAAI,OAAO,EAAE,SAAS,KAAK,SAAS,EAAE,CAAC;QACrC,GAAG,CAAC,GAAG,CAAC,IAAI,CAAC,SAAS,GAAG,OAAO,CAAC,SAAS,CAAC;IAC7C,CAAC;IAED,MAAM,kBAAkB,GAAG,cAAc,EAAE,CAAC,CAAC,CAAC,CAAC,QAAQ,EAAE,MAAM,CAAC,CAAC,CAAC,CAAC,CAAC,MAAM,CAAC,CAAC;IAC5E,MAAM,OAAO,GACX,OAAO,KAAK,KAAK,QAAQ;QACvB,CAAC,CAAC,MAAM,GAAG,CAAC,gBAAgB,CAAC,MAAM,CAAC,KAAK,EAAE,EAAE,kBAAkB,EAAE,CAAC;QAClE,CAAC,CAAC,MAAM,GAAG,CAAC,gBAAgB,CAAC,MAAM,CAC/B,KAAK,YAAY,UAAU,CAAC,CAAC,CAAC,KAAK,CAAC,CAAC,CAAC,IAAI,UAAU,CAAC,KAAK,CAAC,EAC3D,EAAE,kBAAkB,EAAE,CACvB,CAAC;IACR,MAAM,YAAY,GAAG,OAAO,CAAC,UAAU,CAAC,CAAC,CAAC,CAAC;IAC3C,MAAM,aAAa,GAAG,OAAO,CAAC,WAAW,CAAC,CAAC,CAAC,CAAC;IAC7C,IAAI,CAAC,YAAY,IAAI,CAAC,aAAa,EAAE,CAAC;QACpC,MAAM,IAAI,KAAK,CAAC,oDAAoD,CAAC,CAAC;IACxE,CAAC;IACD,MAAM,SAAS,GAAW,YAAY,CAAC;IACvC,MAAM,UAAU,GAAW,aAAa,CAAC;IAEzC,KAAK,UAAU,mBAAmB,CAAC,KAAgB;QACjD,MAAM,MAAM,GAAG,UAAU,CAAC,KAAK,CAAC,CAAC;QACjC,MAAM,OAAO,GAAG,MAAM,OAAO,CAAC,GAAG,CAAC,EAAE,CAAC,SAAS,CAAC,EAAE,MAAM,EAAE,CAAC,CAAC;QAC3D,MAAM,MAAM,GAAG,OAAO,CAAC,UAAU,CAAC,CAAC;QACnC,IAAI,CAAC,MAAM,EAAE,CAAC;YACZ,MAAM,IAAI,KAAK,CAAC,oDAAoD,UAAU,GAAG,CAAC,CAAC;QACrF,CAAC;QAED,MAAM,IAAI,GAAG,aAAa,CAAC,MAAM,CAAC,IAAoB,CAAC,CAAC;QACxD,MAAM,MAAM,GAAG,YAAY,CAAC,IAAI,EAAE,KAAK,CAAC,KAAK,EAAE,KAAK,CAAC,MAAM,CAAC,CAAC;QAC7D,OAAO,EAAE,MAAM,EAAE,KAAK,EAAE,KAAK,CAAC,KAAK,EAAE,MAAM,EAAE,KAAK,CAAC,MAAM,EAAE,CAAC;IAC9D,CAAC;IAED,OAAO;QACL,mBAAmB;QACnB,KAAK,CAAC,gBAAgB,CAAC,KAAgB;YACrC,MAAM,EAAE,MAAM,EAAE,SAAS,EAAE,GAAG,MAAM,mBAAmB,CAAC,KAAK,CAAC,CAAC;YAE/D,MAAM,MAAM,GAAG,eAAe,CAAC,KAAK,CAAC,KAAK,EAAE,KAAK,CAAC,MAAM,CAAC,CAAC;YAC1D,MAAM,OAAO,GAAG,KAAK,CAAC,IAAI,CAAC;YAC3B,MAAM,OAAO,GAAG,MAAM,CAAC,IAAI,CAAC;YAC5B,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,SAAS,CAAC,MAAM,EAAE,CAAC,EAAE,EAAE,CAAC;gBAC1C,MAAM,QAAQ,GAAG,OAAO,CAAC,CAAC,GAAG,CAAC,GAAG,CAAC,CAAC,IAAI,GAAG,CAAC;gBAC3C,OAAO,CAAC,CAAC,GAAG,CAAC,CAAC,GAAG,OAAO,CAAC,CAAC,GAAG,CAAC,CAAC,IAAI,CAAC,CAAC;gBACrC,OAAO,CAAC,CAAC,GAAG,CAAC,GAAG,CAAC,CAAC,GAAG,OAAO,CAAC,CAAC,GAAG,CAAC,GAAG,CAAC,CAAC,IAAI,CAAC,CAAC;gBAC7C,OAAO,CAAC,CAAC,GAAG,CAAC,GAAG,CAAC,CAAC,GAAG,OAAO,CAAC,CAAC,GAAG,CAAC,GAAG,CAAC,CAAC,IAAI,CAAC,CAAC;gBAC7C,OAAO,CAAC,CAAC,GAAG,CAAC,GAAG,CAAC,CAAC,GAAG,IAAI,CAAC,KAAK,CAAC,CAAC,SAAS,CAAC,CAAC,CAAC,IAAI,CAAC,CAAC,GAAG,QAAQ,CAAC,CAAC;YAClE,CAAC;YACD,OAAO,MAAM,CAAC;QAChB,CAAC;KACF,CAAC;AACJ,CAAC"}

package/dist/browser-image-codec.d.ts

@@ -0,0 +1,3 @@
+import type { ImageCodec } from "./onnx-background-remover.js";
+export declare const browserImageCodec: ImageCodec;
+//# sourceMappingURL=browser-image-codec.d.ts.map

package/dist/browser-image-codec.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"browser-image-codec.d.ts","sourceRoot":"","sources":["../src/browser-image-codec.ts"],"names":[],"mappings":"AAKA,OAAO,KAAK,EAAE,UAAU,EAAe,MAAM,8BAA8B,CAAC;AAW5E,eAAO,MAAM,iBAAiB,EAAE,UA4B/B,CAAC"}

package/dist/browser-image-codec.js

@@ -0,0 +1,32 @@
+function createContext(width, height) {
+    const canvas = document.createElement("canvas");
+    canvas.width = width;
+    canvas.height = height;
+    const ctx = canvas.getContext("2d", { willReadFrequently: true });
+    if (!ctx)
+        throw new Error("Canvas is unavailable in this browser.");
+    return ctx;
+}
+export const browserImageCodec = {
+    async encode(image) {
+        const ctx = createContext(image.width, image.height);
+        ctx.putImageData(new ImageData(image.data, image.width, image.height), 0, 0);
+        const blob = await new Promise((resolve) => ctx.canvas.toBlob(resolve, "image/png"));
+        if (!blob)
+            throw new Error("The browser could not encode the image as PNG.");
+        return { mimeType: "image/png", data: new Uint8Array(await blob.arrayBuffer()) };
+    },
+    async decode(image) {
+        const bitmap = await createImageBitmap(new Blob([image.data], { type: image.mimeType }));
+        try {
+            const ctx = createContext(bitmap.width, bitmap.height);
+            ctx.drawImage(bitmap, 0, 0);
+            const { data } = ctx.getImageData(0, 0, bitmap.width, bitmap.height);
+            return { width: bitmap.width, height: bitmap.height, data };
+        }
+        finally {
+            bitmap.close();
+        }
+    },
+};
+//# sourceMappingURL=browser-image-codec.js.map

package/dist/browser-image-codec.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"browser-image-codec.js","sourceRoot":"","sources":["../src/browser-image-codec.ts"],"names":[],"mappings":"AAOA,SAAS,aAAa,CAAC,KAAa,EAAE,MAAc;IAClD,MAAM,MAAM,GAAG,QAAQ,CAAC,aAAa,CAAC,QAAQ,CAAC,CAAC;IAChD,MAAM,CAAC,KAAK,GAAG,KAAK,CAAC;IACrB,MAAM,CAAC,MAAM,GAAG,MAAM,CAAC;IACvB,MAAM,GAAG,GAAG,MAAM,CAAC,UAAU,CAAC,IAAI,EAAE,EAAE,kBAAkB,EAAE,IAAI,EAAE,CAAC,CAAC;IAClE,IAAI,CAAC,GAAG;QAAE,MAAM,IAAI,KAAK,CAAC,wCAAwC,CAAC,CAAC;IACpE,OAAO,GAAG,CAAC;AACb,CAAC;AAED,MAAM,CAAC,MAAM,iBAAiB,GAAe;IAC3C,KAAK,CAAC,MAAM,CAAC,KAAkB;QAC7B,MAAM,GAAG,GAAG,aAAa,CAAC,KAAK,CAAC,KAAK,EAAE,KAAK,CAAC,MAAM,CAAC,CAAC;QACrD,GAAG,CAAC,YAAY,CACd,IAAI,SAAS,CAAC,KAAK,CAAC,IAAsC,EAAE,KAAK,CAAC,KAAK,EAAE,KAAK,CAAC,MAAM,CAAC,EACtF,CAAC,EACD,CAAC,CACF,CAAC;QACF,MAAM,IAAI,GAAG,MAAM,IAAI,OAAO,CAAc,CAAC,OAAO,EAAE,EAAE,CACtD,GAAG,CAAC,MAAM,CAAC,MAAM,CAAC,OAAO,EAAE,WAAW,CAAC,CACxC,CAAC;QACF,IAAI,CAAC,IAAI;YAAE,MAAM,IAAI,KAAK,CAAC,gDAAgD,CAAC,CAAC;QAC7E,OAAO,EAAE,QAAQ,EAAE,WAAW,EAAE,IAAI,EAAE,IAAI,UAAU,CAAC,MAAM,IAAI,CAAC,WAAW,EAAE,CAAC,EAAE,CAAC;IACnF,CAAC;IAED,KAAK,CAAC,MAAM,CAAC,KAAc;QACzB,MAAM,MAAM,GAAG,MAAM,iBAAiB,CACpC,IAAI,IAAI,CAAC,CAAC,KAAK,CAAC,IAA+B,CAAC,EAAE,EAAE,IAAI,EAAE,KAAK,CAAC,QAAQ,EAAE,CAAC,CAC5E,CAAC;QACF,IAAI,CAAC;YACH,MAAM,GAAG,GAAG,aAAa,CAAC,MAAM,CAAC,KAAK,EAAE,MAAM,CAAC,MAAM,CAAC,CAAC;YACvD,GAAG,CAAC,SAAS,CAAC,MAAM,EAAE,CAAC,EAAE,CAAC,CAAC,CAAC;YAC5B,MAAM,EAAE,IAAI,EAAE,GAAG,GAAG,CAAC,YAAY,CAAC,CAAC,EAAE,CAAC,EAAE,MAAM,CAAC,KAAK,EAAE,MAAM,CAAC,MAAM,CAAC,CAAC;YACrE,OAAO,EAAE,KAAK,EAAE,MAAM,CAAC,KAAK,EAAE,MAAM,EAAE,MAAM,CAAC,MAAM,EAAE,IAAI,EAAE,CAAC;QAC9D,CAAC;gBAAS,CAAC;YACT,MAAM,CAAC,KAAK,EAAE,CAAC;QACjB,CAAC;IACH,CAAC;CACF,CAAC"}

package/dist/index.d.ts

@@ -0,0 +1,4 @@
+export * from "./background-removal.js";
+export * from "./upscale.js";
+export * from "./auto-crop.js";
+//# sourceMappingURL=index.d.ts.map

package/dist/index.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"index.d.ts","sourceRoot":"","sources":["../src/index.ts"],"names":[],"mappings":"AAAA,cAAc,yBAAyB,CAAC;AACxC,cAAc,cAAc,CAAC;AAC7B,cAAc,gBAAgB,CAAC"}

package/dist/index.js

@@ -0,0 +1,4 @@
+export * from "./background-removal.js";
+export * from "./upscale.js";
+export * from "./auto-crop.js";
+//# sourceMappingURL=index.js.map

package/dist/index.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"index.js","sourceRoot":"","sources":["../src/index.ts"],"names":[],"mappings":"AAAA,cAAc,yBAAyB,CAAC;AACxC,cAAc,cAAc,CAAC;AAC7B,cAAc,gBAAgB,CAAC"}

package/dist/onnx-background-remover.d.ts

@@ -0,0 +1,49 @@
+/**
+ * The in-browser U²-Net background remover, as an `ImageGenerator`.
+ *
+ * It runs the model from this package (ONNX Runtime Web) on the visitor's own
+ * machine, so it needs no server, no account and no network once the model is
+ * cached — and, being an `ImageGenerator` from
+ * `@richardmcquiston01/ai-providers`, it is interchangeable with any other
+ * provider (a local ComfyUI, a hosted API). Providers exchange *encoded*
+ * images, so an {@link ImageCodec} turns them into pixels and back.
+ */
+import { type AiImage, type ImageGenerator, type ImageRequest, type ImageResult, type Result } from "@richardmcquiston01/ai-providers";
+import type { BackgroundRemovalModel, LoadBackgroundRemovalModelOptions } from "./background-removal.js";
+/** The one thing this provider does. */
+export declare const REMOVE_BACKGROUND_MODE = "remove-background";
+/** RGBA pixels, row-major: structurally an `ImageData`, without needing one. */
+export interface PixelBuffer {
+    width: number;
+    height: number;
+    data: Uint8ClampedArray;
+}
+/** Converts between the encoded images providers exchange and RGBA pixels. */
+export interface ImageCodec {
+    encode(image: PixelBuffer): Promise<AiImage>;
+    decode(image: AiImage): Promise<PixelBuffer>;
+}
+export interface OnnxBackgroundRemoverOptions {
+    /** Loads the model. Called once; a failed load is retried on the next run. */
+    loadModel: () => Promise<BackgroundRemovalModel>;
+    codec: ImageCodec;
+}
+export declare class OnnxBackgroundRemover implements ImageGenerator {
+    readonly provider = "onnx-web";
+    readonly model = "u2netp";
+    private readonly loadModel;
+    private readonly codec;
+    private modelPromise;
+    constructor(options: OnnxBackgroundRemoverOptions);
+    listModes(): Promise<Result<readonly string[]>>;
+    generateImage(request: ImageRequest): Promise<Result<ImageResult>>;
+    /** Load the model once and share it; forget a failed load so it can retry. */
+    private getModel;
+}
+/**
+ * A `loadModel` for {@link OnnxBackgroundRemover} that imports ONNX Runtime
+ * only when the model is first needed, so a bundler keeps it out of the chunk
+ * that merely lists the provider.
+ */
+export declare function lazyBackgroundRemovalModel(model: string | ArrayBuffer | Uint8Array, options?: LoadBackgroundRemovalModelOptions): () => Promise<BackgroundRemovalModel>;
+//# sourceMappingURL=onnx-background-remover.d.ts.map

package/dist/onnx-background-remover.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"onnx-background-remover.d.ts","sourceRoot":"","sources":["../src/onnx-background-remover.ts"],"names":[],"mappings":"AAAA;;;;;;;;;GASG;AACH,OAAO,EAIL,KAAK,OAAO,EACZ,KAAK,cAAc,EACnB,KAAK,YAAY,EACjB,KAAK,WAAW,EAChB,KAAK,MAAM,EACZ,MAAM,kCAAkC,CAAC;AAC1C,OAAO,KAAK,EACV,sBAAsB,EACtB,iCAAiC,EAClC,MAAM,yBAAyB,CAAC;AAEjC,wCAAwC;AACxC,eAAO,MAAM,sBAAsB,sBAAsB,CAAC;AAE1D,gFAAgF;AAChF,MAAM,WAAW,WAAW;IAC1B,KAAK,EAAE,MAAM,CAAC;IACd,MAAM,EAAE,MAAM,CAAC;IACf,IAAI,EAAE,iBAAiB,CAAC;CACzB;AAED,8EAA8E;AAC9E,MAAM,WAAW,UAAU;IACzB,MAAM,CAAC,KAAK,EAAE,WAAW,GAAG,OAAO,CAAC,OAAO,CAAC,CAAC;IAC7C,MAAM,CAAC,KAAK,EAAE,OAAO,GAAG,OAAO,CAAC,WAAW,CAAC,CAAC;CAC9C;AAED,MAAM,WAAW,4BAA4B;IAC3C,8EAA8E;IAC9E,SAAS,EAAE,MAAM,OAAO,CAAC,sBAAsB,CAAC,CAAC;IACjD,KAAK,EAAE,UAAU,CAAC;CACnB;AAED,qBAAa,qBAAsB,YAAW,cAAc;IAC1D,QAAQ,CAAC,QAAQ,cAAc;IAC/B,QAAQ,CAAC,KAAK,YAAY;IAE1B,OAAO,CAAC,QAAQ,CAAC,SAAS,CAAwC;IAClE,OAAO,CAAC,QAAQ,CAAC,KAAK,CAAa;IACnC,OAAO,CAAC,YAAY,CAAgD;gBAExD,OAAO,EAAE,4BAA4B;IAK3C,SAAS,IAAI,OAAO,CAAC,MAAM,CAAC,SAAS,MAAM,EAAE,CAAC,CAAC;IAI/C,aAAa,CAAC,OAAO,EAAE,YAAY,GAAG,OAAO,CAAC,MAAM,CAAC,WAAW,CAAC,CAAC;IAqCxE,8EAA8E;IAC9E,OAAO,CAAC,QAAQ;CASjB;AAED;;;;GAIG;AACH,wBAAgB,0BAA0B,CACxC,KAAK,EAAE,MAAM,GAAG,WAAW,GAAG,UAAU,EACxC,OAAO,CAAC,EAAE,iCAAiC,GAC1C,MAAM,OAAO,CAAC,sBAAsB,CAAC,CAKvC"}

package/dist/onnx-background-remover.js

@@ -0,0 +1,80 @@
+/**
+ * The in-browser U²-Net background remover, as an `ImageGenerator`.
+ *
+ * It runs the model from this package (ONNX Runtime Web) on the visitor's own
+ * machine, so it needs no server, no account and no network once the model is
+ * cached — and, being an `ImageGenerator` from
+ * `@richardmcquiston01/ai-providers`, it is interchangeable with any other
+ * provider (a local ComfyUI, a hosted API). Providers exchange *encoded*
+ * images, so an {@link ImageCodec} turns them into pixels and back.
+ */
+import { describeUnknownError, err, ok, } from "@richardmcquiston01/ai-providers";
+/** The one thing this provider does. */
+export const REMOVE_BACKGROUND_MODE = "remove-background";
+export class OnnxBackgroundRemover {
+    provider = "onnx-web";
+    model = "u2netp";
+    loadModel;
+    codec;
+    modelPromise = null;
+    constructor(options) {
+        this.loadModel = options.loadModel;
+        this.codec = options.codec;
+    }
+    async listModes() {
+        return ok([REMOVE_BACKGROUND_MODE]);
+    }
+    async generateImage(request) {
+        const label = `${this.provider} (model "${this.model}")`;
+        if (request.mode !== null && request.mode !== REMOVE_BACKGROUND_MODE) {
+            return err(`${label} has no mode "${request.mode}"; its only mode is "${REMOVE_BACKGROUND_MODE}".`);
+        }
+        const input = request.images[0];
+        if (!input) {
+            return err(`${label} needs the image to cut out, but the request had none.`);
+        }
+        if (request.signal?.aborted) {
+            return err(`${label} was cancelled before it started.`);
+        }
+        try {
+            const model = await this.getModel();
+            const pixels = await this.codec.decode(input);
+            const cutout = await model.removeBackground(new ImageData(pixels.data, pixels.width, pixels.height));
+            // The model can't be interrupted mid-run, so honour a cancel afterwards.
+            if (request.signal?.aborted) {
+                return err(`${label} was cancelled while it was running.`);
+            }
+            const image = await this.codec.encode({
+                width: cutout.width,
+                height: cutout.height,
+                data: cutout.data,
+            });
+            return ok({ image, notes: null });
+        }
+        catch (cause) {
+            return err(`${label} failed: ${describeUnknownError(cause)}`, cause);
+        }
+    }
+    /** Load the model once and share it; forget a failed load so it can retry. */
+    getModel() {
+        if (!this.modelPromise) {
+            this.modelPromise = this.loadModel().catch((cause) => {
+                this.modelPromise = null;
+                throw cause;
+            });
+        }
+        return this.modelPromise;
+    }
+}
+/**
+ * A `loadModel` for {@link OnnxBackgroundRemover} that imports ONNX Runtime
+ * only when the model is first needed, so a bundler keeps it out of the chunk
+ * that merely lists the provider.
+ */
+export function lazyBackgroundRemovalModel(model, options) {
+    return async () => {
+        const { loadBackgroundRemovalModel } = await import("./background-removal.js");
+        return loadBackgroundRemovalModel(model, options);
+    };
+}
+//# sourceMappingURL=onnx-background-remover.js.map

package/dist/onnx-background-remover.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"onnx-background-remover.js","sourceRoot":"","sources":["../src/onnx-background-remover.ts"],"names":[],"mappings":"AAAA;;;;;;;;;GASG;AACH,OAAO,EACL,oBAAoB,EACpB,GAAG,EACH,EAAE,GAMH,MAAM,kCAAkC,CAAC;AAM1C,wCAAwC;AACxC,MAAM,CAAC,MAAM,sBAAsB,GAAG,mBAAmB,CAAC;AAqB1D,MAAM,OAAO,qBAAqB;IACvB,QAAQ,GAAG,UAAU,CAAC;IACtB,KAAK,GAAG,QAAQ,CAAC;IAET,SAAS,CAAwC;IACjD,KAAK,CAAa;IAC3B,YAAY,GAA2C,IAAI,CAAC;IAEpE,YAAY,OAAqC;QAC/C,IAAI,CAAC,SAAS,GAAG,OAAO,CAAC,SAAS,CAAC;QACnC,IAAI,CAAC,KAAK,GAAG,OAAO,CAAC,KAAK,CAAC;IAC7B,CAAC;IAED,KAAK,CAAC,SAAS;QACb,OAAO,EAAE,CAAC,CAAC,sBAAsB,CAAC,CAAC,CAAC;IACtC,CAAC;IAED,KAAK,CAAC,aAAa,CAAC,OAAqB;QACvC,MAAM,KAAK,GAAG,GAAG,IAAI,CAAC,QAAQ,YAAY,IAAI,CAAC,KAAK,IAAI,CAAC;QAEzD,IAAI,OAAO,CAAC,IAAI,KAAK,IAAI,IAAI,OAAO,CAAC,IAAI,KAAK,sBAAsB,EAAE,CAAC;YACrE,OAAO,GAAG,CACR,GAAG,KAAK,iBAAiB,OAAO,CAAC,IAAI,wBAAwB,sBAAsB,IAAI,CACxF,CAAC;QACJ,CAAC;QACD,MAAM,KAAK,GAAG,OAAO,CAAC,MAAM,CAAC,CAAC,CAAC,CAAC;QAChC,IAAI,CAAC,KAAK,EAAE,CAAC;YACX,OAAO,GAAG,CAAC,GAAG,KAAK,wDAAwD,CAAC,CAAC;QAC/E,CAAC;QACD,IAAI,OAAO,CAAC,MAAM,EAAE,OAAO,EAAE,CAAC;YAC5B,OAAO,GAAG,CAAC,GAAG,KAAK,mCAAmC,CAAC,CAAC;QAC1D,CAAC;QAED,IAAI,CAAC;YACH,MAAM,KAAK,GAAG,MAAM,IAAI,CAAC,QAAQ,EAAE,CAAC;YACpC,MAAM,MAAM,GAAG,MAAM,IAAI,CAAC,KAAK,CAAC,MAAM,CAAC,KAAK,CAAC,CAAC;YAC9C,MAAM,MAAM,GAAG,MAAM,KAAK,CAAC,gBAAgB,CACzC,IAAI,SAAS,CAAC,MAAM,CAAC,IAAsC,EAAE,MAAM,CAAC,KAAK,EAAE,MAAM,CAAC,MAAM,CAAC,CAC1F,CAAC;YACF,yEAAyE;YACzE,IAAI,OAAO,CAAC,MAAM,EAAE,OAAO,EAAE,CAAC;gBAC5B,OAAO,GAAG,CAAC,GAAG,KAAK,sCAAsC,CAAC,CAAC;YAC7D,CAAC;YACD,MAAM,KAAK,GAAG,MAAM,IAAI,CAAC,KAAK,CAAC,MAAM,CAAC;gBACpC,KAAK,EAAE,MAAM,CAAC,KAAK;gBACnB,MAAM,EAAE,MAAM,CAAC,MAAM;gBACrB,IAAI,EAAE,MAAM,CAAC,IAAI;aAClB,CAAC,CAAC;YACH,OAAO,EAAE,CAAC,EAAE,KAAK,EAAE,KAAK,EAAE,IAAI,EAAE,CAAC,CAAC;QACpC,CAAC;QAAC,OAAO,KAAK,EAAE,CAAC;YACf,OAAO,GAAG,CAAC,GAAG,KAAK,YAAY,oBAAoB,CAAC,KAAK,CAAC,EAAE,EAAE,KAAK,CAAC,CAAC;QACvE,CAAC;IACH,CAAC;IAED,8EAA8E;IACtE,QAAQ;QACd,IAAI,CAAC,IAAI,CAAC,YAAY,EAAE,CAAC;YACvB,IAAI,CAAC,YAAY,GAAG,IAAI,CAAC,SAAS,EAAE,CAAC,KAAK,CAAC,CAAC,KAAc,EAAE,EAAE;gBAC5D,IAAI,CAAC,YAAY,GAAG,IAAI,CAAC;gBACzB,MAAM,KAAK,CAAC;YACd,CAAC,CAAC,CAAC;QACL,CAAC;QACD,OAAO,IAAI,CAAC,YAAY,CAAC;IAC3B,CAAC;CACF;AAED;;;;GAIG;AACH,MAAM,UAAU,0BAA0B,CACxC,KAAwC,EACxC,OAA2C;IAE3C,OAAO,KAAK,IAAI,EAAE;QAChB,MAAM,EAAE,0BAA0B,EAAE,GAAG,MAAM,MAAM,CAAC,yBAAyB,CAAC,CAAC;QAC/E,OAAO,0BAA0B,CAAC,KAAK,EAAE,OAAO,CAAC,CAAC;IACpD,CAAC,CAAC;AACJ,CAAC"}

package/dist/provider.d.ts

@@ -0,0 +1,3 @@
+export * from "./onnx-background-remover.js";
+export * from "./browser-image-codec.js";
+//# sourceMappingURL=provider.d.ts.map

package/dist/provider.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"provider.d.ts","sourceRoot":"","sources":["../src/provider.ts"],"names":[],"mappings":"AAGA,cAAc,8BAA8B,CAAC;AAC7C,cAAc,0BAA0B,CAAC"}

package/dist/provider.js

@@ -0,0 +1,6 @@
+// Entry point for `@richardmcquiston01/ai-tools/provider`: the background
+// remover as an `ImageGenerator`. Kept apart from the main entry so that code
+// that only wants the raw models never needs `@richardmcquiston01/ai-providers`.
+export * from "./onnx-background-remover.js";
+export * from "./browser-image-codec.js";
+//# sourceMappingURL=provider.js.map

package/dist/provider.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"provider.js","sourceRoot":"","sources":["../src/provider.ts"],"names":[],"mappings":"AAAA,0EAA0E;AAC1E,8EAA8E;AAC9E,iFAAiF;AACjF,cAAc,8BAA8B,CAAC;AAC7C,cAAc,0BAA0B,CAAC"}

package/dist/upscale.d.ts

@@ -0,0 +1,20 @@
+export type UpscaleFactor = 2 | 3 | 4 | 8;
+export interface LoadUpscaleModelOptions {
+    /**
+     * Full URL to this scale factor's `model.json` (with its weight-shard files
+     * served alongside it), overriding the library's default behavior of
+     * fetching from the jsdelivr/unpkg CDN mirrors of the npm package. Needed
+     * for a fully self-contained deployment with no external CDN dependency —
+     * the caller is expected to serve the vendored model files (bundled from
+     * `@upscalerjs/esrgan-slim`'s own `models/<scale>x/` directory) and pass
+     * their resolved URL here.
+     */
+    modelUrl?: string;
+}
+export interface UpscaleModel {
+    readonly scaleFactor: UpscaleFactor;
+    /** Upscales `image` by this model's scale factor, preserving the alpha channel. */
+    upscale(image: ImageData): Promise<ImageData>;
+}
+export declare function loadUpscaleModel(scaleFactor: UpscaleFactor, options?: LoadUpscaleModelOptions): Promise<UpscaleModel>;
+//# sourceMappingURL=upscale.d.ts.map

package/dist/upscale.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"upscale.d.ts","sourceRoot":"","sources":["../src/upscale.ts"],"names":[],"mappings":"AAQA,MAAM,MAAM,aAAa,GAAG,CAAC,GAAG,CAAC,GAAG,CAAC,GAAG,CAAC,CAAC;AAc1C,MAAM,WAAW,uBAAuB;IACtC;;;;;;;;OAQG;IACH,QAAQ,CAAC,EAAE,MAAM,CAAC;CACnB;AAED,MAAM,WAAW,YAAY;IAC3B,QAAQ,CAAC,WAAW,EAAE,aAAa,CAAC;IACpC,mFAAmF;IACnF,OAAO,CAAC,KAAK,EAAE,SAAS,GAAG,OAAO,CAAC,SAAS,CAAC,CAAC;CAC/C;AAkCD,wBAAsB,gBAAgB,CACpC,WAAW,EAAE,aAAa,EAC1B,OAAO,CAAC,EAAE,uBAAuB,GAChC,OAAO,CAAC,YAAY,CAAC,CAyCvB"}

package/dist/upscale.js

@@ -0,0 +1,83 @@
+import { createImageData, resize } from "@richardmcquiston01/core-image";
+import * as tf from "@tensorflow/tfjs";
+import x2 from "@upscalerjs/esrgan-slim/2x";
+import x3 from "@upscalerjs/esrgan-slim/3x";
+import x4 from "@upscalerjs/esrgan-slim/4x";
+import x8 from "@upscalerjs/esrgan-slim/8x";
+import Upscaler from "upscaler";
+const MODEL_DEFINITIONS = {
+    2: x2,
+    3: x3,
+    4: x4,
+    8: x8,
+};
+function imageToRgbTensor(image) {
+    const pixelCount = image.width * image.height;
+    const rgb = new Int32Array(pixelCount * 3);
+    const data = image.data;
+    for (let i = 0; i < pixelCount; i++) {
+        rgb[i * 3] = data[i * 4] ?? 0;
+        rgb[i * 3 + 1] = data[i * 4 + 1] ?? 0;
+        rgb[i * 3 + 2] = data[i * 4 + 2] ?? 0;
+    }
+    return tf.tensor3d(rgb, [image.height, image.width, 3], "int32");
+}
+/** Upsamples the alpha channel to the output size by round-tripping it through the RGBA resize filter (the upscaling model only understands RGB). */
+function upsampleAlpha(image, outWidth, outHeight) {
+    const alphaImage = createImageData(image.width, image.height);
+    const src = image.data;
+    const dst = alphaImage.data;
+    for (let i = 0; i < image.width * image.height; i++) {
+        const a = src[i * 4 + 3] ?? 255;
+        dst[i * 4] = a;
+        dst[i * 4 + 1] = a;
+        dst[i * 4 + 2] = a;
+        dst[i * 4 + 3] = 255;
+    }
+    const resized = resize(alphaImage, { width: outWidth, height: outHeight, method: "bilinear" });
+    const result = new Uint8ClampedArray(outWidth * outHeight);
+    for (let i = 0; i < result.length; i++) {
+        result[i] = resized.data[i * 4] ?? 255;
+    }
+    return result;
+}
+export async function loadUpscaleModel(scaleFactor, options) {
+    const baseDefinition = MODEL_DEFINITIONS[scaleFactor];
+    const modelDefinition = options?.modelUrl
+        ? { ...baseDefinition, path: options.modelUrl }
+        : baseDefinition;
+    const upscaler = new Upscaler({ model: modelDefinition });
+    await upscaler.ready;
+    return {
+        scaleFactor,
+        async upscale(image) {
+            const inputTensor = imageToRgbTensor(image);
+            let outputTensor;
+            try {
+                outputTensor = await upscaler.upscale(inputTensor, { output: "tensor" });
+            }
+            finally {
+                inputTensor.dispose();
+            }
+            const [outHeight, outWidth] = outputTensor.shape;
+            if (outHeight === undefined || outWidth === undefined) {
+                outputTensor.dispose();
+                throw new Error("Upscale model produced a tensor with an unexpected shape");
+            }
+            const pixels = await outputTensor.data();
+            outputTensor.dispose();
+            const alpha = upsampleAlpha(image, outWidth, outHeight);
+            const result = createImageData(outWidth, outHeight);
+            const dst = result.data;
+            const pixelCount = outWidth * outHeight;
+            for (let i = 0; i < pixelCount; i++) {
+                dst[i * 4] = pixels[i * 3] ?? 0;
+                dst[i * 4 + 1] = pixels[i * 3 + 1] ?? 0;
+                dst[i * 4 + 2] = pixels[i * 3 + 2] ?? 0;
+                dst[i * 4 + 3] = alpha[i] ?? 255;
+            }
+            return result;
+        },
+    };
+}
+//# sourceMappingURL=upscale.js.map

package/dist/upscale.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"upscale.js","sourceRoot":"","sources":["../src/upscale.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,eAAe,EAAE,MAAM,EAAE,MAAM,gCAAgC,CAAC;AACzE,OAAO,KAAK,EAAE,MAAM,kBAAkB,CAAC;AACvC,OAAO,EAAE,MAAM,4BAA4B,CAAC;AAC5C,OAAO,EAAE,MAAM,4BAA4B,CAAC;AAC5C,OAAO,EAAE,MAAM,4BAA4B,CAAC;AAC5C,OAAO,EAAE,MAAM,4BAA4B,CAAC;AAC5C,OAAO,QAAQ,MAAM,UAAU,CAAC;AAShC,MAAM,iBAAiB,GAAmD;IACxE,CAAC,EAAE,EAA6B;IAChC,CAAC,EAAE,EAA6B;IAChC,CAAC,EAAE,EAA6B;IAChC,CAAC,EAAE,EAA6B;CACjC,CAAC;AAqBF,SAAS,gBAAgB,CAAC,KAAgB;IACxC,MAAM,UAAU,GAAG,KAAK,CAAC,KAAK,GAAG,KAAK,CAAC,MAAM,CAAC;IAC9C,MAAM,GAAG,GAAG,IAAI,UAAU,CAAC,UAAU,GAAG,CAAC,CAAC,CAAC;IAC3C,MAAM,IAAI,GAAG,KAAK,CAAC,IAAI,CAAC;IACxB,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,UAAU,EAAE,CAAC,EAAE,EAAE,CAAC;QACpC,GAAG,CAAC,CAAC,GAAG,CAAC,CAAC,GAAG,IAAI,CAAC,CAAC,GAAG,CAAC,CAAC,IAAI,CAAC,CAAC;QAC9B,GAAG,CAAC,CAAC,GAAG,CAAC,GAAG,CAAC,CAAC,GAAG,IAAI,CAAC,CAAC,GAAG,CAAC,GAAG,CAAC,CAAC,IAAI,CAAC,CAAC;QACtC,GAAG,CAAC,CAAC,GAAG,CAAC,GAAG,CAAC,CAAC,GAAG,IAAI,CAAC,CAAC,GAAG,CAAC,GAAG,CAAC,CAAC,IAAI,CAAC,CAAC;IACxC,CAAC;IACD,OAAO,EAAE,CAAC,QAAQ,CAAC,GAAG,EAAE,CAAC,KAAK,CAAC,MAAM,EAAE,KAAK,CAAC,KAAK,EAAE,CAAC,CAAC,EAAE,OAAO,CAAC,CAAC;AACnE,CAAC;AAED,qJAAqJ;AACrJ,SAAS,aAAa,CAAC,KAAgB,EAAE,QAAgB,EAAE,SAAiB;IAC1E,MAAM,UAAU,GAAG,eAAe,CAAC,KAAK,CAAC,KAAK,EAAE,KAAK,CAAC,MAAM,CAAC,CAAC;IAC9D,MAAM,GAAG,GAAG,KAAK,CAAC,IAAI,CAAC;IACvB,MAAM,GAAG,GAAG,UAAU,CAAC,IAAI,CAAC;IAC5B,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,KAAK,CAAC,KAAK,GAAG,KAAK,CAAC,MAAM,EAAE,CAAC,EAAE,EAAE,CAAC;QACpD,MAAM,CAAC,GAAG,GAAG,CAAC,CAAC,GAAG,CAAC,GAAG,CAAC,CAAC,IAAI,GAAG,CAAC;QAChC,GAAG,CAAC,CAAC,GAAG,CAAC,CAAC,GAAG,CAAC,CAAC;QACf,GAAG,CAAC,CAAC,GAAG,CAAC,GAAG,CAAC,CAAC,GAAG,CAAC,CAAC;QACnB,GAAG,CAAC,CAAC,GAAG,CAAC,GAAG,CAAC,CAAC,GAAG,CAAC,CAAC;QACnB,GAAG,CAAC,CAAC,GAAG,CAAC,GAAG,CAAC,CAAC,GAAG,GAAG,CAAC;IACvB,CAAC;IACD,MAAM,OAAO,GAAG,MAAM,CAAC,UAAU,EAAE,EAAE,KAAK,EAAE,QAAQ,EAAE,MAAM,EAAE,SAAS,EAAE,MAAM,EAAE,UAAU,EAAE,CAAC,CAAC;IAC/F,MAAM,MAAM,GAAG,IAAI,iBAAiB,CAAC,QAAQ,GAAG,SAAS,CAAC,CAAC;IAC3D,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,MAAM,CAAC,MAAM,EAAE,CAAC,EAAE,EAAE,CAAC;QACvC,MAAM,CAAC,CAAC,CAAC,GAAG,OAAO,CAAC,IAAI,CAAC,CAAC,GAAG,CAAC,CAAC,IAAI,GAAG,CAAC;IACzC,CAAC;IACD,OAAO,MAAM,CAAC;AAChB,CAAC;AAED,MAAM,CAAC,KAAK,UAAU,gBAAgB,CACpC,WAA0B,EAC1B,OAAiC;IAEjC,MAAM,cAAc,GAAG,iBAAiB,CAAC,WAAW,CAAC,CAAC;IACtD,MAAM,eAAe,GAA4B,OAAO,EAAE,QAAQ;QAChE,CAAC,CAAC,EAAE,GAAG,cAAc,EAAE,IAAI,EAAE,OAAO,CAAC,QAAQ,EAAE;QAC/C,CAAC,CAAC,cAAc,CAAC;IACnB,MAAM,QAAQ,GAAG,IAAI,QAAQ,CAAC,EAAE,KAAK,EAAE,eAAe,EAAE,CAAC,CAAC;IAC1D,MAAM,QAAQ,CAAC,KAAK,CAAC;IAErB,OAAO;QACL,WAAW;QACX,KAAK,CAAC,OAAO,CAAC,KAAgB;YAC5B,MAAM,WAAW,GAAG,gBAAgB,CAAC,KAAK,CAAC,CAAC;YAC5C,IAAI,YAAyB,CAAC;YAC9B,IAAI,CAAC;gBACH,YAAY,GAAG,MAAM,QAAQ,CAAC,OAAO,CAAC,WAAW,EAAE,EAAE,MAAM,EAAE,QAAQ,EAAE,CAAC,CAAC;YAC3E,CAAC;oBAAS,CAAC;gBACT,WAAW,CAAC,OAAO,EAAE,CAAC;YACxB,CAAC;YAED,MAAM,CAAC,SAAS,EAAE,QAAQ,CAAC,GAAG,YAAY,CAAC,KAAK,CAAC;YACjD,IAAI,SAAS,KAAK,SAAS,IAAI,QAAQ,KAAK,SAAS,EAAE,CAAC;gBACtD,YAAY,CAAC,OAAO,EAAE,CAAC;gBACvB,MAAM,IAAI,KAAK,CAAC,0DAA0D,CAAC,CAAC;YAC9E,CAAC;YAED,MAAM,MAAM,GAAG,MAAM,YAAY,CAAC,IAAI,EAAE,CAAC;YACzC,YAAY,CAAC,OAAO,EAAE,CAAC;YAEvB,MAAM,KAAK,GAAG,aAAa,CAAC,KAAK,EAAE,QAAQ,EAAE,SAAS,CAAC,CAAC;YACxD,MAAM,MAAM,GAAG,eAAe,CAAC,QAAQ,EAAE,SAAS,CAAC,CAAC;YACpD,MAAM,GAAG,GAAG,MAAM,CAAC,IAAI,CAAC;YACxB,MAAM,UAAU,GAAG,QAAQ,GAAG,SAAS,CAAC;YACxC,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,UAAU,EAAE,CAAC,EAAE,EAAE,CAAC;gBACpC,GAAG,CAAC,CAAC,GAAG,CAAC,CAAC,GAAG,MAAM,CAAC,CAAC,GAAG,CAAC,CAAC,IAAI,CAAC,CAAC;gBAChC,GAAG,CAAC,CAAC,GAAG,CAAC,GAAG,CAAC,CAAC,GAAG,MAAM,CAAC,CAAC,GAAG,CAAC,GAAG,CAAC,CAAC,IAAI,CAAC,CAAC;gBACxC,GAAG,CAAC,CAAC,GAAG,CAAC,GAAG,CAAC,CAAC,GAAG,MAAM,CAAC,CAAC,GAAG,CAAC,GAAG,CAAC,CAAC,IAAI,CAAC,CAAC;gBACxC,GAAG,CAAC,CAAC,GAAG,CAAC,GAAG,CAAC,CAAC,GAAG,KAAK,CAAC,CAAC,CAAC,IAAI,GAAG,CAAC;YACnC,CAAC;YACD,OAAO,MAAM,CAAC;QAChB,CAAC;KACF,CAAC;AACJ,CAAC"}

package/models/LICENSE-APACHE-2.0.txt

@@ -0,0 +1,202 @@
+
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright [yyyy] [name of copyright owner]
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.

package/models/NOTICE.md

@@ -0,0 +1,18 @@
+# Model attribution
+
+## `u2netp.onnx`
+
+- **Architecture**: U²-Net (small variant, "U2NETP"), from Qin et al., _"U^2-Net: Going
+  Deeper with Nested U-Structure for Salient Object Detection"_, Pattern Recognition 2020.
+- **Source repository**: https://github.com/xuebinqin/U-2-Net
+- **License**: Apache License 2.0 — full text in `LICENSE-APACHE-2.0.txt` alongside this file (the u2net/u2netp checkpoints — NOT the separate
+  `u2net_portrait` checkpoint, which carries a non-commercial restriction and is not
+  used here).
+- **File obtained from**: the pre-exported ONNX release published by
+  https://github.com/danielgatis/rembg (MIT-licensed project; this file is an
+  unmodified redistribution of the original Apache-2.0 licensed weights), at
+  https://github.com/danielgatis/rembg/releases/download/v0.0.0/u2netp.onnx
+  (SHA-256: `309c8469258dda742793dce0ebea8e6dd393174f89934733ecc8b14c76f4ddd8`).
+
+Vendored directly into this package so background removal runs fully offline with
+no runtime dependency on any external CDN or model-hosting service.

package/package.json

@@ -1,6 +1,76 @@
 {
   "name": "@richardmcquiston01/ai-tools",
-  "version": "0.0.0-stage",
-  "stub": true,
-  "description": "Temporary package placeholder for staged publishing"
-}
+  "version": "0.2.0",
+  "description": "In-browser AI image tools (background removal, upscaling, auto-crop) for laser/CNC/maker workflows",
+  "keywords": [
+    "background-removal",
+    "upscaling",
+    "auto-crop",
+    "onnx",
+    "image-processing",
+    "laser",
+    "cnc",
+    "engraving",
+    "maker",
+    "typescript",
+    "image-generator"
+  ],
+  "license": "MIT",
+  "homepage": "https://github.com/RichardMcQuiston01/maker-image-tools#readme",
+  "bugs": {
+    "url": "https://github.com/RichardMcQuiston01/maker-image-tools/issues"
+  },
+  "type": "module",
+  "repository": {
+    "type": "git",
+    "url": "git+https://github.com/RichardMcQuiston01/maker-image-tools.git",
+    "directory": "packages/ai-tools"
+  },
+  "main": "./dist/index.js",
+  "types": "./dist/index.d.ts",
+  "exports": {
+    ".": {
+      "types": "./dist/index.d.ts",
+      "default": "./dist/index.js"
+    },
+    "./provider": {
+      "types": "./dist/provider.d.ts",
+      "default": "./dist/provider.js"
+    }
+  },
+  "files": [
+    "dist",
+    "models",
+    "README.md",
+    "LICENSE"
+  ],
+  "publishConfig": {
+    "access": "public"
+  },
+  "scripts": {
+    "build": "tsc -p tsconfig.json",
+    "test": "vitest run",
+    "typecheck": "tsc -p tsconfig.json --noEmit"
+  },
+  "dependencies": {
+    "@richardmcquiston01/core-image": "0.2.0",
+    "@tensorflow/tfjs": "4.11.0",
+    "@upscalerjs/esrgan-slim": "1.0.0",
+    "onnxruntime-web": "1.27.0",
+    "upscaler": "1.0.0"
+  },
+  "devDependencies": {
+    "@richardmcquiston01/ai-providers": "^0.1.0",
+    "jsdom": "^25.0.1",
+    "typescript": "^5.7.3",
+    "vitest": "^3.0.4"
+  },
+  "peerDependencies": {
+    "@richardmcquiston01/ai-providers": "^0.1.0"
+  },
+  "peerDependenciesMeta": {
+    "@richardmcquiston01/ai-providers": {
+      "optional": true
+    }
+  }
+}