mirror of
https://github.com/snapotter-hq/SnapOtter.git
synced 2026-08-03 07:46:42 +02:00
feat: add Phase 4 AI tools with Python bridge and 6 new tools
Add Python bridge (packages/ai/src/bridge.ts) that calls Python scripts via child_process with venv-first fallback to system python3. Implements 6 AI-powered tools: - Remove Background: rembg-based with U2-Net/IS-Net models - Image Upscaling: Real-ESRGAN with Lanczos fallback - OCR/Text Extraction: Tesseract + PaddleOCR engines - Face/PII Blur: MediaPipe face detection with configurable blur - Object Eraser: LaMa inpainting with mask-based input - Smart Crop: Sharp attention-based entropy cropping (no Python needed) Each tool includes: Python script, TypeScript wrapper, API route, and React settings component. All Python scripts handle ImportError gracefully with clear installation messages.
This commit is contained in:
Siddharth Kumar Sah
@@ -17,6 +17,7 @@
|
||||
"@fastify/static": "^8.1.0",
|
||||
"@fastify/swagger": "^9.4.0",
|
||||
"@fastify/swagger-ui": "^5.2.0",
|
||||
"@stirling-image/ai": "workspace:*",
|
||||
"@stirling-image/image-engine": "workspace:*",
|
||||
"@stirling-image/shared": "workspace:*",
|
||||
"archiver": "^7.0.1",
|
||||
|
||||
@@ -0,0 +1,86 @@
|
||||
import type { FastifyInstance, FastifyRequest, FastifyReply } from "fastify";
|
||||
import { randomUUID } from "node:crypto";
|
||||
import { writeFile } from "node:fs/promises";
|
||||
import { join, basename } from "node:path";
|
||||
import { blurFaces } from "@stirling-image/ai";
|
||||
import { createWorkspace } from "../../lib/workspace.js";
|
||||
|
||||
/**
|
||||
* Face detection and blurring route.
|
||||
* Uses MediaPipe for detection, PIL for blurring.
|
||||
*/
|
||||
export function registerBlurFaces(app: FastifyInstance) {
|
||||
app.post(
|
||||
"/api/v1/tools/blur-faces",
|
||||
async (request: FastifyRequest, reply: FastifyReply) => {
|
||||
let fileBuffer: Buffer | null = null;
|
||||
let filename = "image";
|
||||
let settingsRaw: string | null = null;
|
||||
|
||||
try {
|
||||
const parts = request.parts();
|
||||
for await (const part of parts) {
|
||||
if (part.type === "file") {
|
||||
const chunks: Buffer[] = [];
|
||||
for await (const chunk of part.file) {
|
||||
chunks.push(chunk);
|
||||
}
|
||||
fileBuffer = Buffer.concat(chunks);
|
||||
filename = basename(part.filename ?? "image");
|
||||
} else if (part.fieldname === "settings") {
|
||||
settingsRaw = part.value as string;
|
||||
}
|
||||
}
|
||||
} catch (err) {
|
||||
return reply.status(400).send({
|
||||
error: "Failed to parse multipart request",
|
||||
details: err instanceof Error ? err.message : String(err),
|
||||
});
|
||||
}
|
||||
|
||||
if (!fileBuffer || fileBuffer.length === 0) {
|
||||
return reply.status(400).send({ error: "No image file provided" });
|
||||
}
|
||||
|
||||
try {
|
||||
const settings = settingsRaw ? JSON.parse(settingsRaw) : {};
|
||||
const jobId = randomUUID();
|
||||
const workspacePath = await createWorkspace(jobId);
|
||||
|
||||
// Save input
|
||||
const inputPath = join(workspacePath, "input", filename);
|
||||
await writeFile(inputPath, fileBuffer);
|
||||
|
||||
// Process
|
||||
const result = await blurFaces(
|
||||
fileBuffer,
|
||||
join(workspacePath, "output"),
|
||||
{
|
||||
blurRadius: settings.blurRadius ?? 30,
|
||||
sensitivity: settings.sensitivity ?? 0.5,
|
||||
},
|
||||
);
|
||||
|
||||
// Save output
|
||||
const outputFilename =
|
||||
filename.replace(/\.[^.]+$/, "") + "_blurred.png";
|
||||
const outputPath = join(workspacePath, "output", outputFilename);
|
||||
await writeFile(outputPath, result.buffer);
|
||||
|
||||
return reply.send({
|
||||
jobId,
|
||||
downloadUrl: `/api/v1/download/${jobId}/${encodeURIComponent(outputFilename)}`,
|
||||
originalSize: fileBuffer.length,
|
||||
processedSize: result.buffer.length,
|
||||
facesDetected: result.facesDetected,
|
||||
faces: result.faces,
|
||||
});
|
||||
} catch (err) {
|
||||
return reply.status(422).send({
|
||||
error: "Face blur failed",
|
||||
details: err instanceof Error ? err.message : "Unknown error",
|
||||
});
|
||||
}
|
||||
},
|
||||
);
|
||||
}
|
||||
@@ -0,0 +1,88 @@
|
||||
import type { FastifyInstance, FastifyRequest, FastifyReply } from "fastify";
|
||||
import { randomUUID } from "node:crypto";
|
||||
import { writeFile } from "node:fs/promises";
|
||||
import { join, basename } from "node:path";
|
||||
import { inpaint } from "@stirling-image/ai";
|
||||
import { createWorkspace } from "../../lib/workspace.js";
|
||||
|
||||
/**
|
||||
* Object eraser / inpainting route.
|
||||
* Accepts an image and a mask image, erases masked areas.
|
||||
*/
|
||||
export function registerEraseObject(app: FastifyInstance) {
|
||||
app.post(
|
||||
"/api/v1/tools/erase-object",
|
||||
async (request: FastifyRequest, reply: FastifyReply) => {
|
||||
let imageBuffer: Buffer | null = null;
|
||||
let maskBuffer: Buffer | null = null;
|
||||
let filename = "image";
|
||||
|
||||
try {
|
||||
const parts = request.parts();
|
||||
for await (const part of parts) {
|
||||
if (part.type === "file") {
|
||||
const chunks: Buffer[] = [];
|
||||
for await (const chunk of part.file) {
|
||||
chunks.push(chunk);
|
||||
}
|
||||
const buf = Buffer.concat(chunks);
|
||||
if (part.fieldname === "mask") {
|
||||
maskBuffer = buf;
|
||||
} else {
|
||||
imageBuffer = buf;
|
||||
filename = basename(part.filename ?? "image");
|
||||
}
|
||||
}
|
||||
}
|
||||
} catch (err) {
|
||||
return reply.status(400).send({
|
||||
error: "Failed to parse multipart request",
|
||||
details: err instanceof Error ? err.message : String(err),
|
||||
});
|
||||
}
|
||||
|
||||
if (!imageBuffer || imageBuffer.length === 0) {
|
||||
return reply.status(400).send({ error: "No image file provided" });
|
||||
}
|
||||
if (!maskBuffer || maskBuffer.length === 0) {
|
||||
return reply
|
||||
.status(400)
|
||||
.send({ error: "No mask image provided. Upload a mask as a second file with fieldname 'mask'" });
|
||||
}
|
||||
|
||||
try {
|
||||
const jobId = randomUUID();
|
||||
const workspacePath = await createWorkspace(jobId);
|
||||
|
||||
// Save input
|
||||
const inputPath = join(workspacePath, "input", filename);
|
||||
await writeFile(inputPath, imageBuffer);
|
||||
|
||||
// Process
|
||||
const resultBuffer = await inpaint(
|
||||
imageBuffer,
|
||||
maskBuffer,
|
||||
join(workspacePath, "output"),
|
||||
);
|
||||
|
||||
// Save output
|
||||
const outputFilename =
|
||||
filename.replace(/\.[^.]+$/, "") + "_erased.png";
|
||||
const outputPath = join(workspacePath, "output", outputFilename);
|
||||
await writeFile(outputPath, resultBuffer);
|
||||
|
||||
return reply.send({
|
||||
jobId,
|
||||
downloadUrl: `/api/v1/download/${jobId}/${encodeURIComponent(outputFilename)}`,
|
||||
originalSize: imageBuffer.length,
|
||||
processedSize: resultBuffer.length,
|
||||
});
|
||||
} catch (err) {
|
||||
return reply.status(422).send({
|
||||
error: "Object erasing failed",
|
||||
details: err instanceof Error ? err.message : "Unknown error",
|
||||
});
|
||||
}
|
||||
},
|
||||
);
|
||||
}
|
||||
@@ -32,6 +32,13 @@ import { registerFavicon } from "./favicon.js";
|
||||
import { registerImageToPdf } from "./image-to-pdf.js";
|
||||
// Phase 3: Adjustments extra
|
||||
import { registerReplaceColor } from "./replace-color.js";
|
||||
// Phase 4: AI Tools
|
||||
import { registerRemoveBackground } from "./remove-background.js";
|
||||
import { registerUpscale } from "./upscale.js";
|
||||
import { registerOcr } from "./ocr.js";
|
||||
import { registerBlurFaces } from "./blur-faces.js";
|
||||
import { registerEraseObject } from "./erase-object.js";
|
||||
import { registerSmartCrop } from "./smart-crop.js";
|
||||
|
||||
/**
|
||||
* Registry that imports and registers all tool routes.
|
||||
@@ -79,5 +86,13 @@ export async function registerToolRoutes(app: FastifyInstance): Promise<void> {
|
||||
// Phase 3: Adjustments extra
|
||||
registerReplaceColor(app);
|
||||
|
||||
app.log.info("Tool routes registered (26 tools, 29 endpoints)");
|
||||
// Phase 4: AI Tools
|
||||
registerRemoveBackground(app);
|
||||
registerUpscale(app);
|
||||
registerOcr(app);
|
||||
registerBlurFaces(app);
|
||||
registerEraseObject(app);
|
||||
registerSmartCrop(app);
|
||||
|
||||
app.log.info("Tool routes registered (32 tools, 35 endpoints)");
|
||||
}
|
||||
|
||||
@@ -0,0 +1,68 @@
|
||||
import type { FastifyInstance, FastifyRequest, FastifyReply } from "fastify";
|
||||
import { randomUUID } from "node:crypto";
|
||||
import { basename } from "node:path";
|
||||
import { extractText } from "@stirling-image/ai";
|
||||
import { createWorkspace } from "../../lib/workspace.js";
|
||||
|
||||
/**
|
||||
* OCR / text extraction route.
|
||||
* Returns JSON with extracted text rather than an image.
|
||||
*/
|
||||
export function registerOcr(app: FastifyInstance) {
|
||||
app.post(
|
||||
"/api/v1/tools/ocr",
|
||||
async (request: FastifyRequest, reply: FastifyReply) => {
|
||||
let fileBuffer: Buffer | null = null;
|
||||
let filename = "image";
|
||||
let settingsRaw: string | null = null;
|
||||
|
||||
try {
|
||||
const parts = request.parts();
|
||||
for await (const part of parts) {
|
||||
if (part.type === "file") {
|
||||
const chunks: Buffer[] = [];
|
||||
for await (const chunk of part.file) {
|
||||
chunks.push(chunk);
|
||||
}
|
||||
fileBuffer = Buffer.concat(chunks);
|
||||
filename = basename(part.filename ?? "image");
|
||||
} else if (part.fieldname === "settings") {
|
||||
settingsRaw = part.value as string;
|
||||
}
|
||||
}
|
||||
} catch (err) {
|
||||
return reply.status(400).send({
|
||||
error: "Failed to parse multipart request",
|
||||
details: err instanceof Error ? err.message : String(err),
|
||||
});
|
||||
}
|
||||
|
||||
if (!fileBuffer || fileBuffer.length === 0) {
|
||||
return reply.status(400).send({ error: "No image file provided" });
|
||||
}
|
||||
|
||||
try {
|
||||
const settings = settingsRaw ? JSON.parse(settingsRaw) : {};
|
||||
const jobId = randomUUID();
|
||||
const workspacePath = await createWorkspace(jobId);
|
||||
|
||||
const result = await extractText(fileBuffer, workspacePath, {
|
||||
engine: settings.engine,
|
||||
language: settings.language,
|
||||
});
|
||||
|
||||
return reply.send({
|
||||
jobId,
|
||||
filename,
|
||||
text: result.text,
|
||||
engine: result.engine,
|
||||
});
|
||||
} catch (err) {
|
||||
return reply.status(422).send({
|
||||
error: "OCR failed",
|
||||
details: err instanceof Error ? err.message : "Unknown error",
|
||||
});
|
||||
}
|
||||
},
|
||||
);
|
||||
}
|
||||
@@ -0,0 +1,80 @@
|
||||
import type { FastifyInstance, FastifyRequest, FastifyReply } from "fastify";
|
||||
import { randomUUID } from "node:crypto";
|
||||
import { writeFile } from "node:fs/promises";
|
||||
import { join, basename } from "node:path";
|
||||
import { removeBackground } from "@stirling-image/ai";
|
||||
import { createWorkspace } from "../../lib/workspace.js";
|
||||
|
||||
/**
|
||||
* AI background removal route.
|
||||
* Uses Python + rembg under the hood.
|
||||
*/
|
||||
export function registerRemoveBackground(app: FastifyInstance) {
|
||||
app.post(
|
||||
"/api/v1/tools/remove-background",
|
||||
async (request: FastifyRequest, reply: FastifyReply) => {
|
||||
let fileBuffer: Buffer | null = null;
|
||||
let filename = "image";
|
||||
let settingsRaw: string | null = null;
|
||||
|
||||
try {
|
||||
const parts = request.parts();
|
||||
for await (const part of parts) {
|
||||
if (part.type === "file") {
|
||||
const chunks: Buffer[] = [];
|
||||
for await (const chunk of part.file) {
|
||||
chunks.push(chunk);
|
||||
}
|
||||
fileBuffer = Buffer.concat(chunks);
|
||||
filename = basename(part.filename ?? "image");
|
||||
} else if (part.fieldname === "settings") {
|
||||
settingsRaw = part.value as string;
|
||||
}
|
||||
}
|
||||
} catch (err) {
|
||||
return reply.status(400).send({
|
||||
error: "Failed to parse multipart request",
|
||||
details: err instanceof Error ? err.message : String(err),
|
||||
});
|
||||
}
|
||||
|
||||
if (!fileBuffer || fileBuffer.length === 0) {
|
||||
return reply.status(400).send({ error: "No image file provided" });
|
||||
}
|
||||
|
||||
try {
|
||||
const settings = settingsRaw ? JSON.parse(settingsRaw) : {};
|
||||
const jobId = randomUUID();
|
||||
const workspacePath = await createWorkspace(jobId);
|
||||
|
||||
// Save input
|
||||
const inputPath = join(workspacePath, "input", filename);
|
||||
await writeFile(inputPath, fileBuffer);
|
||||
|
||||
// Process
|
||||
const resultBuffer = await removeBackground(
|
||||
fileBuffer,
|
||||
join(workspacePath, "output"),
|
||||
{ model: settings.model },
|
||||
);
|
||||
|
||||
// Save output
|
||||
const outputFilename = filename.replace(/\.[^.]+$/, "") + "_nobg.png";
|
||||
const outputPath = join(workspacePath, "output", outputFilename);
|
||||
await writeFile(outputPath, resultBuffer);
|
||||
|
||||
return reply.send({
|
||||
jobId,
|
||||
downloadUrl: `/api/v1/download/${jobId}/${encodeURIComponent(outputFilename)}`,
|
||||
originalSize: fileBuffer.length,
|
||||
processedSize: resultBuffer.length,
|
||||
});
|
||||
} catch (err) {
|
||||
return reply.status(422).send({
|
||||
error: "Background removal failed",
|
||||
details: err instanceof Error ? err.message : "Unknown error",
|
||||
});
|
||||
}
|
||||
},
|
||||
);
|
||||
}
|
||||
@@ -0,0 +1,33 @@
|
||||
import { z } from "zod";
|
||||
import sharp from "sharp";
|
||||
import type { FastifyInstance } from "fastify";
|
||||
import { createToolRoute } from "../tool-factory.js";
|
||||
|
||||
const settingsSchema = z.object({
|
||||
width: z.number().int().positive(),
|
||||
height: z.number().int().positive(),
|
||||
});
|
||||
|
||||
/**
|
||||
* Smart crop using Sharp's attention-based strategy.
|
||||
* Uses entropy/saliency detection to find the most interesting region.
|
||||
* No Python needed.
|
||||
*/
|
||||
export function registerSmartCrop(app: FastifyInstance) {
|
||||
createToolRoute(app, {
|
||||
toolId: "smart-crop",
|
||||
settingsSchema,
|
||||
process: async (inputBuffer, settings, filename) => {
|
||||
const result = await sharp(inputBuffer)
|
||||
.resize(settings.width, settings.height, {
|
||||
fit: "cover",
|
||||
position: sharp.strategy.attention,
|
||||
})
|
||||
.png()
|
||||
.toBuffer();
|
||||
|
||||
const outputFilename = filename.replace(/\.[^.]+$/, "") + "_smartcrop.png";
|
||||
return { buffer: result, filename: outputFilename, contentType: "image/png" };
|
||||
},
|
||||
});
|
||||
}
|
||||
@@ -0,0 +1,86 @@
|
||||
import type { FastifyInstance, FastifyRequest, FastifyReply } from "fastify";
|
||||
import { randomUUID } from "node:crypto";
|
||||
import { writeFile } from "node:fs/promises";
|
||||
import { join, basename } from "node:path";
|
||||
import { upscale } from "@stirling-image/ai";
|
||||
import { createWorkspace } from "../../lib/workspace.js";
|
||||
|
||||
/**
|
||||
* AI image upscaling route.
|
||||
* Uses Real-ESRGAN when available, falls back to Lanczos.
|
||||
*/
|
||||
export function registerUpscale(app: FastifyInstance) {
|
||||
app.post(
|
||||
"/api/v1/tools/upscale",
|
||||
async (request: FastifyRequest, reply: FastifyReply) => {
|
||||
let fileBuffer: Buffer | null = null;
|
||||
let filename = "image";
|
||||
let settingsRaw: string | null = null;
|
||||
|
||||
try {
|
||||
const parts = request.parts();
|
||||
for await (const part of parts) {
|
||||
if (part.type === "file") {
|
||||
const chunks: Buffer[] = [];
|
||||
for await (const chunk of part.file) {
|
||||
chunks.push(chunk);
|
||||
}
|
||||
fileBuffer = Buffer.concat(chunks);
|
||||
filename = basename(part.filename ?? "image");
|
||||
} else if (part.fieldname === "settings") {
|
||||
settingsRaw = part.value as string;
|
||||
}
|
||||
}
|
||||
} catch (err) {
|
||||
return reply.status(400).send({
|
||||
error: "Failed to parse multipart request",
|
||||
details: err instanceof Error ? err.message : String(err),
|
||||
});
|
||||
}
|
||||
|
||||
if (!fileBuffer || fileBuffer.length === 0) {
|
||||
return reply.status(400).send({ error: "No image file provided" });
|
||||
}
|
||||
|
||||
try {
|
||||
const settings = settingsRaw ? JSON.parse(settingsRaw) : {};
|
||||
const scale = Number(settings.scale) || 2;
|
||||
|
||||
const jobId = randomUUID();
|
||||
const workspacePath = await createWorkspace(jobId);
|
||||
|
||||
// Save input
|
||||
const inputPath = join(workspacePath, "input", filename);
|
||||
await writeFile(inputPath, fileBuffer);
|
||||
|
||||
// Process
|
||||
const result = await upscale(
|
||||
fileBuffer,
|
||||
join(workspacePath, "output"),
|
||||
{ scale },
|
||||
);
|
||||
|
||||
// Save output
|
||||
const outputFilename =
|
||||
filename.replace(/\.[^.]+$/, "") + `_${scale}x.png`;
|
||||
const outputPath = join(workspacePath, "output", outputFilename);
|
||||
await writeFile(outputPath, result.buffer);
|
||||
|
||||
return reply.send({
|
||||
jobId,
|
||||
downloadUrl: `/api/v1/download/${jobId}/${encodeURIComponent(outputFilename)}`,
|
||||
originalSize: fileBuffer.length,
|
||||
processedSize: result.buffer.length,
|
||||
width: result.width,
|
||||
height: result.height,
|
||||
method: result.method,
|
||||
});
|
||||
} catch (err) {
|
||||
return reply.status(422).send({
|
||||
error: "Upscaling failed",
|
||||
details: err instanceof Error ? err.message : "Unknown error",
|
||||
});
|
||||
}
|
||||
},
|
||||
);
|
||||
}
|
||||
@@ -0,0 +1,104 @@
|
||||
import { useState } from "react";
|
||||
import { useFileStore } from "@/stores/file-store";
|
||||
import { useToolProcessor } from "@/hooks/use-tool-processor";
|
||||
import { Download, Loader2 } from "lucide-react";
|
||||
|
||||
export function BlurFacesSettings() {
|
||||
const { files } = useFileStore();
|
||||
const { processFiles, processing, error, downloadUrl, originalSize, processedSize } =
|
||||
useToolProcessor("blur-faces");
|
||||
|
||||
const [blurRadius, setBlurRadius] = useState(30);
|
||||
const [sensitivity, setSensitivity] = useState(50);
|
||||
|
||||
const handleProcess = () => {
|
||||
processFiles(files, {
|
||||
blurRadius,
|
||||
sensitivity: sensitivity / 100,
|
||||
});
|
||||
};
|
||||
|
||||
const hasFile = files.length > 0;
|
||||
|
||||
return (
|
||||
<div className="space-y-4">
|
||||
{/* Blur radius */}
|
||||
<div>
|
||||
<div className="flex justify-between items-center">
|
||||
<label className="text-xs text-muted-foreground">Blur Radius</label>
|
||||
<span className="text-xs font-mono text-foreground">{blurRadius}</span>
|
||||
</div>
|
||||
<input
|
||||
type="range"
|
||||
min={5}
|
||||
max={80}
|
||||
value={blurRadius}
|
||||
onChange={(e) => setBlurRadius(Number(e.target.value))}
|
||||
className="w-full mt-1"
|
||||
/>
|
||||
<div className="flex justify-between text-[10px] text-muted-foreground mt-0.5">
|
||||
<span>Light</span>
|
||||
<span>Heavy</span>
|
||||
</div>
|
||||
</div>
|
||||
|
||||
{/* Sensitivity */}
|
||||
<div>
|
||||
<div className="flex justify-between items-center">
|
||||
<label className="text-xs text-muted-foreground">Detection Sensitivity</label>
|
||||
<span className="text-xs font-mono text-foreground">{sensitivity}%</span>
|
||||
</div>
|
||||
<input
|
||||
type="range"
|
||||
min={10}
|
||||
max={90}
|
||||
value={sensitivity}
|
||||
onChange={(e) => setSensitivity(Number(e.target.value))}
|
||||
className="w-full mt-1"
|
||||
/>
|
||||
<div className="flex justify-between text-[10px] text-muted-foreground mt-0.5">
|
||||
<span>More faces</span>
|
||||
<span>Fewer false positives</span>
|
||||
</div>
|
||||
</div>
|
||||
|
||||
{/* Info */}
|
||||
<p className="text-[10px] text-muted-foreground">
|
||||
Uses MediaPipe for face detection. Automatically detects and blurs all faces in the image.
|
||||
</p>
|
||||
|
||||
{/* Error */}
|
||||
{error && <p className="text-xs text-red-500">{error}</p>}
|
||||
|
||||
{/* Size info */}
|
||||
{originalSize != null && processedSize != null && (
|
||||
<div className="text-xs text-muted-foreground space-y-0.5">
|
||||
<p>Original: {(originalSize / 1024).toFixed(1)} KB</p>
|
||||
<p>Processed: {(processedSize / 1024).toFixed(1)} KB</p>
|
||||
</div>
|
||||
)}
|
||||
|
||||
{/* Process button */}
|
||||
<button
|
||||
onClick={handleProcess}
|
||||
disabled={!hasFile || processing}
|
||||
className="w-full py-2.5 rounded-lg bg-primary text-primary-foreground font-medium disabled:opacity-50 disabled:cursor-not-allowed flex items-center justify-center gap-2"
|
||||
>
|
||||
{processing && <Loader2 className="h-4 w-4 animate-spin" />}
|
||||
{processing ? "Detecting Faces..." : "Blur Faces"}
|
||||
</button>
|
||||
|
||||
{/* Download */}
|
||||
{downloadUrl && (
|
||||
<a
|
||||
href={downloadUrl}
|
||||
download
|
||||
className="w-full py-2.5 rounded-lg border border-primary text-primary font-medium flex items-center justify-center gap-2 hover:bg-primary/5"
|
||||
>
|
||||
<Download className="h-4 w-4" />
|
||||
Download
|
||||
</a>
|
||||
)}
|
||||
</div>
|
||||
);
|
||||
}
|
||||
@@ -0,0 +1,125 @@
|
||||
import { useState } from "react";
|
||||
import { useFileStore } from "@/stores/file-store";
|
||||
import { Download, Loader2, Upload } from "lucide-react";
|
||||
|
||||
function getToken(): string {
|
||||
return localStorage.getItem("stirling-token") || "";
|
||||
}
|
||||
|
||||
export function EraseObjectSettings() {
|
||||
const { files, processing, error, setProcessing, setError } = useFileStore();
|
||||
|
||||
const [maskFile, setMaskFile] = useState<File | null>(null);
|
||||
const [downloadUrl, setDownloadUrl] = useState<string | null>(null);
|
||||
const [originalSize, setOriginalSize] = useState<number | null>(null);
|
||||
const [processedSize, setProcessedSize] = useState<number | null>(null);
|
||||
|
||||
const handleMaskSelect = (e: React.ChangeEvent<HTMLInputElement>) => {
|
||||
const selected = e.target.files?.[0];
|
||||
if (selected) setMaskFile(selected);
|
||||
};
|
||||
|
||||
const handleProcess = async () => {
|
||||
if (files.length === 0 || !maskFile) return;
|
||||
|
||||
setProcessing(true);
|
||||
setError(null);
|
||||
setDownloadUrl(null);
|
||||
|
||||
try {
|
||||
const formData = new FormData();
|
||||
formData.append("file", files[0]);
|
||||
formData.append("mask", maskFile);
|
||||
|
||||
const res = await fetch("/api/v1/tools/erase-object", {
|
||||
method: "POST",
|
||||
headers: { Authorization: `Bearer ${getToken()}` },
|
||||
body: formData,
|
||||
});
|
||||
|
||||
if (!res.ok) {
|
||||
const body = await res.json().catch(() => ({}));
|
||||
throw new Error(body.error || body.details || `Failed: ${res.status}`);
|
||||
}
|
||||
|
||||
const data = await res.json();
|
||||
setDownloadUrl(data.downloadUrl);
|
||||
setOriginalSize(data.originalSize);
|
||||
setProcessedSize(data.processedSize);
|
||||
} catch (err) {
|
||||
setError(err instanceof Error ? err.message : "Object erasing failed");
|
||||
} finally {
|
||||
setProcessing(false);
|
||||
}
|
||||
};
|
||||
|
||||
const hasFile = files.length > 0;
|
||||
|
||||
return (
|
||||
<div className="space-y-4">
|
||||
{/* Mask upload */}
|
||||
<div>
|
||||
<label className="text-sm font-medium text-muted-foreground">Mask Image</label>
|
||||
<p className="text-[10px] text-muted-foreground mt-0.5 mb-1.5">
|
||||
Upload a black & white mask where white areas will be erased. Create the mask in any image editor.
|
||||
</p>
|
||||
<label className="flex items-center gap-2 px-3 py-2 rounded border border-dashed border-border cursor-pointer hover:border-primary">
|
||||
<Upload className="h-4 w-4 text-muted-foreground" />
|
||||
<span className="text-xs text-muted-foreground">
|
||||
{maskFile ? maskFile.name : "Select mask image..."}
|
||||
</span>
|
||||
<input
|
||||
type="file"
|
||||
accept="image/*"
|
||||
onChange={handleMaskSelect}
|
||||
className="hidden"
|
||||
/>
|
||||
</label>
|
||||
</div>
|
||||
|
||||
{/* Info */}
|
||||
<div className="p-2 rounded bg-muted text-[10px] text-muted-foreground space-y-1">
|
||||
<p>How to create a mask:</p>
|
||||
<ol className="list-decimal list-inside space-y-0.5">
|
||||
<li>Open your image in any editor</li>
|
||||
<li>Paint white over areas to erase</li>
|
||||
<li>Keep the rest black</li>
|
||||
<li>Export as PNG and upload here</li>
|
||||
</ol>
|
||||
</div>
|
||||
|
||||
{/* Error */}
|
||||
{error && <p className="text-xs text-red-500">{error}</p>}
|
||||
|
||||
{/* Size info */}
|
||||
{originalSize != null && processedSize != null && (
|
||||
<div className="text-xs text-muted-foreground space-y-0.5">
|
||||
<p>Original: {(originalSize / 1024).toFixed(1)} KB</p>
|
||||
<p>Processed: {(processedSize / 1024).toFixed(1)} KB</p>
|
||||
</div>
|
||||
)}
|
||||
|
||||
{/* Process button */}
|
||||
<button
|
||||
onClick={handleProcess}
|
||||
disabled={!hasFile || !maskFile || processing}
|
||||
className="w-full py-2.5 rounded-lg bg-primary text-primary-foreground font-medium disabled:opacity-50 disabled:cursor-not-allowed flex items-center justify-center gap-2"
|
||||
>
|
||||
{processing && <Loader2 className="h-4 w-4 animate-spin" />}
|
||||
{processing ? "Erasing..." : "Erase Object"}
|
||||
</button>
|
||||
|
||||
{/* Download */}
|
||||
{downloadUrl && (
|
||||
<a
|
||||
href={downloadUrl}
|
||||
download
|
||||
className="w-full py-2.5 rounded-lg border border-primary text-primary font-medium flex items-center justify-center gap-2 hover:bg-primary/5"
|
||||
>
|
||||
<Download className="h-4 w-4" />
|
||||
Download
|
||||
</a>
|
||||
)}
|
||||
</div>
|
||||
);
|
||||
}
|
||||
@@ -0,0 +1,165 @@
|
||||
import { useState } from "react";
|
||||
import { useFileStore } from "@/stores/file-store";
|
||||
import { Loader2, Copy, Check } from "lucide-react";
|
||||
|
||||
function getToken(): string {
|
||||
return localStorage.getItem("stirling-token") || "";
|
||||
}
|
||||
|
||||
type OcrEngine = "tesseract" | "paddleocr";
|
||||
|
||||
const LANGUAGES = [
|
||||
{ code: "en", label: "English" },
|
||||
{ code: "de", label: "German" },
|
||||
{ code: "fr", label: "French" },
|
||||
{ code: "es", label: "Spanish" },
|
||||
{ code: "zh", label: "Chinese" },
|
||||
{ code: "ja", label: "Japanese" },
|
||||
{ code: "ko", label: "Korean" },
|
||||
];
|
||||
|
||||
export function OcrSettings() {
|
||||
const { files, processing, error, setProcessing, setError } = useFileStore();
|
||||
|
||||
const [engine, setEngine] = useState<OcrEngine>("tesseract");
|
||||
const [language, setLanguage] = useState("en");
|
||||
const [text, setText] = useState<string | null>(null);
|
||||
const [detectedEngine, setDetectedEngine] = useState<string>("");
|
||||
const [copied, setCopied] = useState(false);
|
||||
|
||||
const handleProcess = async () => {
|
||||
if (files.length === 0) return;
|
||||
|
||||
setProcessing(true);
|
||||
setError(null);
|
||||
setText(null);
|
||||
|
||||
try {
|
||||
const formData = new FormData();
|
||||
formData.append("file", files[0]);
|
||||
formData.append("settings", JSON.stringify({ engine, language }));
|
||||
|
||||
const res = await fetch("/api/v1/tools/ocr", {
|
||||
method: "POST",
|
||||
headers: { Authorization: `Bearer ${getToken()}` },
|
||||
body: formData,
|
||||
});
|
||||
|
||||
if (!res.ok) {
|
||||
const body = await res.json().catch(() => ({}));
|
||||
throw new Error(body.error || body.details || `Failed: ${res.status}`);
|
||||
}
|
||||
|
||||
const data = await res.json();
|
||||
setText(data.text || "");
|
||||
setDetectedEngine(data.engine || engine);
|
||||
} catch (err) {
|
||||
setError(err instanceof Error ? err.message : "OCR failed");
|
||||
} finally {
|
||||
setProcessing(false);
|
||||
}
|
||||
};
|
||||
|
||||
const handleCopy = async () => {
|
||||
if (text) {
|
||||
await navigator.clipboard.writeText(text);
|
||||
setCopied(true);
|
||||
setTimeout(() => setCopied(false), 2000);
|
||||
}
|
||||
};
|
||||
|
||||
const hasFile = files.length > 0;
|
||||
|
||||
return (
|
||||
<div className="space-y-4">
|
||||
{/* Engine selector */}
|
||||
<div>
|
||||
<label className="text-sm font-medium text-muted-foreground">OCR Engine</label>
|
||||
<div className="flex gap-1 mt-1">
|
||||
<button
|
||||
onClick={() => setEngine("tesseract")}
|
||||
className={`flex-1 text-xs py-1.5 rounded ${
|
||||
engine === "tesseract"
|
||||
? "bg-primary text-primary-foreground"
|
||||
: "bg-muted text-muted-foreground"
|
||||
}`}
|
||||
>
|
||||
Tesseract
|
||||
</button>
|
||||
<button
|
||||
onClick={() => setEngine("paddleocr")}
|
||||
className={`flex-1 text-xs py-1.5 rounded ${
|
||||
engine === "paddleocr"
|
||||
? "bg-primary text-primary-foreground"
|
||||
: "bg-muted text-muted-foreground"
|
||||
}`}
|
||||
>
|
||||
PaddleOCR
|
||||
</button>
|
||||
</div>
|
||||
</div>
|
||||
|
||||
{/* Language selector */}
|
||||
<div>
|
||||
<label className="text-xs text-muted-foreground">Language</label>
|
||||
<select
|
||||
value={language}
|
||||
onChange={(e) => setLanguage(e.target.value)}
|
||||
className="w-full mt-0.5 px-2 py-1.5 rounded border border-border bg-background text-sm text-foreground"
|
||||
>
|
||||
{LANGUAGES.map((lang) => (
|
||||
<option key={lang.code} value={lang.code}>
|
||||
{lang.label}
|
||||
</option>
|
||||
))}
|
||||
</select>
|
||||
</div>
|
||||
|
||||
{/* Error */}
|
||||
{error && <p className="text-xs text-red-500">{error}</p>}
|
||||
|
||||
{/* Process button */}
|
||||
<button
|
||||
onClick={handleProcess}
|
||||
disabled={!hasFile || processing}
|
||||
className="w-full py-2.5 rounded-lg bg-primary text-primary-foreground font-medium disabled:opacity-50 disabled:cursor-not-allowed flex items-center justify-center gap-2"
|
||||
>
|
||||
{processing && <Loader2 className="h-4 w-4 animate-spin" />}
|
||||
{processing ? "Extracting Text..." : "Extract Text"}
|
||||
</button>
|
||||
|
||||
{/* Result */}
|
||||
{text !== null && (
|
||||
<div className="space-y-2">
|
||||
<div className="flex items-center justify-between">
|
||||
<label className="text-xs font-medium text-muted-foreground">
|
||||
Extracted Text ({detectedEngine})
|
||||
</label>
|
||||
<button
|
||||
onClick={handleCopy}
|
||||
className="flex items-center gap-1 text-xs text-muted-foreground hover:text-foreground"
|
||||
>
|
||||
{copied ? (
|
||||
<Check className="h-3 w-3" />
|
||||
) : (
|
||||
<Copy className="h-3 w-3" />
|
||||
)}
|
||||
{copied ? "Copied" : "Copy"}
|
||||
</button>
|
||||
</div>
|
||||
<textarea
|
||||
readOnly
|
||||
value={text}
|
||||
rows={8}
|
||||
className="w-full px-2 py-1.5 rounded border border-border bg-muted text-xs text-foreground font-mono resize-y"
|
||||
/>
|
||||
{text.length > 0 && (
|
||||
<p className="text-[10px] text-muted-foreground">
|
||||
{text.length} characters extracted
|
||||
</p>
|
||||
)}
|
||||
</div>
|
||||
)}
|
||||
</div>
|
||||
);
|
||||
}
|
||||
@@ -0,0 +1,108 @@
|
||||
import { useState } from "react";
|
||||
import { useFileStore } from "@/stores/file-store";
|
||||
import { useToolProcessor } from "@/hooks/use-tool-processor";
|
||||
import { Download, Loader2 } from "lucide-react";
|
||||
|
||||
type BgModel = "u2net" | "isnet";
|
||||
|
||||
export function RemoveBgSettings() {
|
||||
const { files } = useFileStore();
|
||||
const { processFiles, processing, error, downloadUrl, originalSize, processedSize } =
|
||||
useToolProcessor("remove-background");
|
||||
|
||||
const [model, setModel] = useState<BgModel>("u2net");
|
||||
const [bgColor, setBgColor] = useState("");
|
||||
|
||||
const handleProcess = () => {
|
||||
const settings: Record<string, unknown> = { model };
|
||||
if (bgColor) settings.backgroundColor = bgColor;
|
||||
processFiles(files, settings);
|
||||
};
|
||||
|
||||
const hasFile = files.length > 0;
|
||||
|
||||
return (
|
||||
<div className="space-y-4">
|
||||
{/* Model selector */}
|
||||
<div>
|
||||
<label className="text-sm font-medium text-muted-foreground">AI Model</label>
|
||||
<select
|
||||
value={model}
|
||||
onChange={(e) => setModel(e.target.value as BgModel)}
|
||||
className="w-full mt-0.5 px-2 py-1.5 rounded border border-border bg-background text-sm text-foreground"
|
||||
>
|
||||
<option value="u2net">U2-Net (General purpose)</option>
|
||||
<option value="isnet">IS-Net (Higher accuracy)</option>
|
||||
</select>
|
||||
</div>
|
||||
|
||||
{/* Background color */}
|
||||
<div>
|
||||
<label className="text-xs text-muted-foreground">
|
||||
Replacement Background (leave empty for transparent)
|
||||
</label>
|
||||
<div className="flex gap-2 mt-0.5">
|
||||
<input
|
||||
type="color"
|
||||
value={bgColor || "#ffffff"}
|
||||
onChange={(e) => setBgColor(e.target.value)}
|
||||
className="w-10 h-8 rounded border border-border cursor-pointer"
|
||||
/>
|
||||
<input
|
||||
type="text"
|
||||
value={bgColor}
|
||||
onChange={(e) => setBgColor(e.target.value)}
|
||||
placeholder="Transparent"
|
||||
className="flex-1 px-2 py-1.5 rounded border border-border bg-background text-sm text-foreground"
|
||||
/>
|
||||
{bgColor && (
|
||||
<button
|
||||
onClick={() => setBgColor("")}
|
||||
className="text-xs text-muted-foreground hover:text-foreground"
|
||||
>
|
||||
Clear
|
||||
</button>
|
||||
)}
|
||||
</div>
|
||||
</div>
|
||||
|
||||
{/* Info */}
|
||||
<p className="text-[10px] text-muted-foreground">
|
||||
Requires Python with rembg installed. Works best with photos of people, products, and animals.
|
||||
</p>
|
||||
|
||||
{/* Error */}
|
||||
{error && <p className="text-xs text-red-500">{error}</p>}
|
||||
|
||||
{/* Size info */}
|
||||
{originalSize != null && processedSize != null && (
|
||||
<div className="text-xs text-muted-foreground space-y-0.5">
|
||||
<p>Original: {(originalSize / 1024).toFixed(1)} KB</p>
|
||||
<p>Processed: {(processedSize / 1024).toFixed(1)} KB</p>
|
||||
</div>
|
||||
)}
|
||||
|
||||
{/* Process button */}
|
||||
<button
|
||||
onClick={handleProcess}
|
||||
disabled={!hasFile || processing}
|
||||
className="w-full py-2.5 rounded-lg bg-primary text-primary-foreground font-medium disabled:opacity-50 disabled:cursor-not-allowed flex items-center justify-center gap-2"
|
||||
>
|
||||
{processing && <Loader2 className="h-4 w-4 animate-spin" />}
|
||||
{processing ? "Removing Background..." : "Remove Background"}
|
||||
</button>
|
||||
|
||||
{/* Download */}
|
||||
{downloadUrl && (
|
||||
<a
|
||||
href={downloadUrl}
|
||||
download
|
||||
className="w-full py-2.5 rounded-lg border border-primary text-primary font-medium flex items-center justify-center gap-2 hover:bg-primary/5"
|
||||
>
|
||||
<Download className="h-4 w-4" />
|
||||
Download
|
||||
</a>
|
||||
)}
|
||||
</div>
|
||||
);
|
||||
}
|
||||
@@ -0,0 +1,131 @@
|
||||
import { useState } from "react";
|
||||
import { useFileStore } from "@/stores/file-store";
|
||||
import { useToolProcessor } from "@/hooks/use-tool-processor";
|
||||
import { Download, Loader2 } from "lucide-react";
|
||||
|
||||
const ASPECT_PRESETS = [
|
||||
{ label: "1:1 Square", w: 1080, h: 1080 },
|
||||
{ label: "16:9 Landscape", w: 1920, h: 1080 },
|
||||
{ label: "9:16 Portrait", w: 1080, h: 1920 },
|
||||
{ label: "4:3 Standard", w: 1440, h: 1080 },
|
||||
{ label: "3:2 Photo", w: 1620, h: 1080 },
|
||||
{ label: "Custom", w: 0, h: 0 },
|
||||
];
|
||||
|
||||
export function SmartCropSettings() {
|
||||
const { files } = useFileStore();
|
||||
const { processFiles, processing, error, downloadUrl, originalSize, processedSize } =
|
||||
useToolProcessor("smart-crop");
|
||||
|
||||
const [width, setWidth] = useState("1080");
|
||||
const [height, setHeight] = useState("1080");
|
||||
const [preset, setPreset] = useState("1:1 Square");
|
||||
|
||||
const handlePreset = (label: string) => {
|
||||
setPreset(label);
|
||||
const p = ASPECT_PRESETS.find((a) => a.label === label);
|
||||
if (p && p.w > 0) {
|
||||
setWidth(String(p.w));
|
||||
setHeight(String(p.h));
|
||||
}
|
||||
};
|
||||
|
||||
const handleProcess = () => {
|
||||
const w = Number(width);
|
||||
const h = Number(height);
|
||||
if (w > 0 && h > 0) {
|
||||
processFiles(files, { width: w, height: h });
|
||||
}
|
||||
};
|
||||
|
||||
const hasFile = files.length > 0;
|
||||
const canProcess = Number(width) > 0 && Number(height) > 0;
|
||||
|
||||
return (
|
||||
<div className="space-y-4">
|
||||
{/* Aspect ratio preset */}
|
||||
<div>
|
||||
<label className="text-sm font-medium text-muted-foreground">Target Aspect Ratio</label>
|
||||
<select
|
||||
value={preset}
|
||||
onChange={(e) => handlePreset(e.target.value)}
|
||||
className="w-full mt-0.5 px-2 py-1.5 rounded border border-border bg-background text-sm text-foreground"
|
||||
>
|
||||
{ASPECT_PRESETS.map((p) => (
|
||||
<option key={p.label} value={p.label}>
|
||||
{p.label}
|
||||
</option>
|
||||
))}
|
||||
</select>
|
||||
</div>
|
||||
|
||||
{/* Width / Height */}
|
||||
<div className="flex gap-2">
|
||||
<div className="flex-1">
|
||||
<label className="text-xs text-muted-foreground">Width (px)</label>
|
||||
<input
|
||||
type="number"
|
||||
value={width}
|
||||
onChange={(e) => {
|
||||
setWidth(e.target.value);
|
||||
setPreset("Custom");
|
||||
}}
|
||||
min={1}
|
||||
className="w-full mt-0.5 px-2 py-1.5 rounded border border-border bg-background text-sm text-foreground"
|
||||
/>
|
||||
</div>
|
||||
<div className="flex-1">
|
||||
<label className="text-xs text-muted-foreground">Height (px)</label>
|
||||
<input
|
||||
type="number"
|
||||
value={height}
|
||||
onChange={(e) => {
|
||||
setHeight(e.target.value);
|
||||
setPreset("Custom");
|
||||
}}
|
||||
min={1}
|
||||
className="w-full mt-0.5 px-2 py-1.5 rounded border border-border bg-background text-sm text-foreground"
|
||||
/>
|
||||
</div>
|
||||
</div>
|
||||
|
||||
{/* Info */}
|
||||
<p className="text-[10px] text-muted-foreground">
|
||||
Uses entropy-based attention detection to find the most interesting region of the image and crops to it.
|
||||
</p>
|
||||
|
||||
{/* Error */}
|
||||
{error && <p className="text-xs text-red-500">{error}</p>}
|
||||
|
||||
{/* Size info */}
|
||||
{originalSize != null && processedSize != null && (
|
||||
<div className="text-xs text-muted-foreground space-y-0.5">
|
||||
<p>Original: {(originalSize / 1024).toFixed(1)} KB</p>
|
||||
<p>Cropped: {(processedSize / 1024).toFixed(1)} KB</p>
|
||||
</div>
|
||||
)}
|
||||
|
||||
{/* Process button */}
|
||||
<button
|
||||
onClick={handleProcess}
|
||||
disabled={!hasFile || !canProcess || processing}
|
||||
className="w-full py-2.5 rounded-lg bg-primary text-primary-foreground font-medium disabled:opacity-50 disabled:cursor-not-allowed flex items-center justify-center gap-2"
|
||||
>
|
||||
{processing && <Loader2 className="h-4 w-4 animate-spin" />}
|
||||
{processing ? "Smart Cropping..." : "Smart Crop"}
|
||||
</button>
|
||||
|
||||
{/* Download */}
|
||||
{downloadUrl && (
|
||||
<a
|
||||
href={downloadUrl}
|
||||
download
|
||||
className="w-full py-2.5 rounded-lg border border-primary text-primary font-medium flex items-center justify-center gap-2 hover:bg-primary/5"
|
||||
>
|
||||
<Download className="h-4 w-4" />
|
||||
Download
|
||||
</a>
|
||||
)}
|
||||
</div>
|
||||
);
|
||||
}
|
||||
@@ -0,0 +1,80 @@
|
||||
import { useState } from "react";
|
||||
import { useFileStore } from "@/stores/file-store";
|
||||
import { useToolProcessor } from "@/hooks/use-tool-processor";
|
||||
import { Download, Loader2 } from "lucide-react";
|
||||
|
||||
export function UpscaleSettings() {
|
||||
const { files } = useFileStore();
|
||||
const { processFiles, processing, error, downloadUrl, originalSize, processedSize } =
|
||||
useToolProcessor("upscale");
|
||||
|
||||
const [scale, setScale] = useState(2);
|
||||
|
||||
const handleProcess = () => {
|
||||
processFiles(files, { scale });
|
||||
};
|
||||
|
||||
const hasFile = files.length > 0;
|
||||
|
||||
return (
|
||||
<div className="space-y-4">
|
||||
{/* Scale factor */}
|
||||
<div>
|
||||
<label className="text-sm font-medium text-muted-foreground">Scale Factor</label>
|
||||
<div className="flex gap-1 mt-1">
|
||||
{[2, 4].map((s) => (
|
||||
<button
|
||||
key={s}
|
||||
onClick={() => setScale(s)}
|
||||
className={`flex-1 text-xs py-1.5 rounded ${
|
||||
scale === s
|
||||
? "bg-primary text-primary-foreground"
|
||||
: "bg-muted text-muted-foreground"
|
||||
}`}
|
||||
>
|
||||
{s}x
|
||||
</button>
|
||||
))}
|
||||
</div>
|
||||
</div>
|
||||
|
||||
{/* Info */}
|
||||
<p className="text-[10px] text-muted-foreground">
|
||||
Uses Real-ESRGAN for AI upscaling when available, otherwise falls back to high-quality Lanczos interpolation.
|
||||
</p>
|
||||
|
||||
{/* Error */}
|
||||
{error && <p className="text-xs text-red-500">{error}</p>}
|
||||
|
||||
{/* Size info */}
|
||||
{originalSize != null && processedSize != null && (
|
||||
<div className="text-xs text-muted-foreground space-y-0.5">
|
||||
<p>Original: {(originalSize / 1024).toFixed(1)} KB</p>
|
||||
<p>Upscaled: {(processedSize / 1024).toFixed(1)} KB</p>
|
||||
</div>
|
||||
)}
|
||||
|
||||
{/* Process button */}
|
||||
<button
|
||||
onClick={handleProcess}
|
||||
disabled={!hasFile || processing}
|
||||
className="w-full py-2.5 rounded-lg bg-primary text-primary-foreground font-medium disabled:opacity-50 disabled:cursor-not-allowed flex items-center justify-center gap-2"
|
||||
>
|
||||
{processing && <Loader2 className="h-4 w-4 animate-spin" />}
|
||||
{processing ? "Upscaling..." : `Upscale ${scale}x`}
|
||||
</button>
|
||||
|
||||
{/* Download */}
|
||||
{downloadUrl && (
|
||||
<a
|
||||
href={downloadUrl}
|
||||
download
|
||||
className="w-full py-2.5 rounded-lg border border-primary text-primary font-medium flex items-center justify-center gap-2 hover:bg-primary/5"
|
||||
>
|
||||
<Download className="h-4 w-4" />
|
||||
Download
|
||||
</a>
|
||||
)}
|
||||
</div>
|
||||
);
|
||||
}
|
||||
@@ -38,6 +38,13 @@ import { FaviconSettings } from "@/components/tools/favicon-settings";
|
||||
import { ImageToPdfSettings } from "@/components/tools/image-to-pdf-settings";
|
||||
// Phase 3: Adjustments extra
|
||||
import { ReplaceColorSettings } from "@/components/tools/replace-color-settings";
|
||||
// Phase 4: AI Tools
|
||||
import { RemoveBgSettings } from "@/components/tools/remove-bg-settings";
|
||||
import { UpscaleSettings } from "@/components/tools/upscale-settings";
|
||||
import { OcrSettings } from "@/components/tools/ocr-settings";
|
||||
import { BlurFacesSettings } from "@/components/tools/blur-faces-settings";
|
||||
import { EraseObjectSettings } from "@/components/tools/erase-object-settings";
|
||||
import { SmartCropSettings } from "@/components/tools/smart-crop-settings";
|
||||
import * as icons from "lucide-react";
|
||||
|
||||
const COLOR_TOOL_IDS = new Set([
|
||||
@@ -85,6 +92,13 @@ function ToolSettingsPanel({ toolId }: { toolId: string }) {
|
||||
if (toolId === "image-to-pdf") return <ImageToPdfSettings />;
|
||||
// Phase 3: Adjustments extra
|
||||
if (toolId === "replace-color") return <ReplaceColorSettings />;
|
||||
// Phase 4: AI Tools
|
||||
if (toolId === "remove-background") return <RemoveBgSettings />;
|
||||
if (toolId === "upscale") return <UpscaleSettings />;
|
||||
if (toolId === "ocr") return <OcrSettings />;
|
||||
if (toolId === "blur-faces") return <BlurFacesSettings />;
|
||||
if (toolId === "erase-object") return <EraseObjectSettings />;
|
||||
if (toolId === "smart-crop") return <SmartCropSettings />;
|
||||
|
||||
return (
|
||||
<p className="text-xs text-muted-foreground italic">
|
||||
|
||||
@@ -10,9 +10,11 @@
|
||||
"clean": "rm -rf dist"
|
||||
},
|
||||
"dependencies": {
|
||||
"@stirling-image/shared": "workspace:*"
|
||||
"@stirling-image/shared": "workspace:*",
|
||||
"sharp": "^0.33.0"
|
||||
},
|
||||
"devDependencies": {
|
||||
"@types/node": "^22.0.0",
|
||||
"typescript": "^5.7.0"
|
||||
}
|
||||
}
|
||||
|
||||
@@ -0,0 +1,95 @@
|
||||
"""Face detection and blurring using MediaPipe."""
|
||||
import sys
|
||||
import json
|
||||
|
||||
|
||||
def main():
|
||||
input_path = sys.argv[1]
|
||||
output_path = sys.argv[2]
|
||||
settings = json.loads(sys.argv[3]) if len(sys.argv) > 3 else {}
|
||||
|
||||
blur_radius = settings.get("blurRadius", 30)
|
||||
sensitivity = settings.get("sensitivity", 0.5)
|
||||
|
||||
try:
|
||||
from PIL import Image, ImageFilter
|
||||
|
||||
img = Image.open(input_path).convert("RGB")
|
||||
|
||||
try:
|
||||
import mediapipe as mp
|
||||
import numpy as np
|
||||
|
||||
mp_face = mp.solutions.face_detection
|
||||
|
||||
with mp_face.FaceDetection(
|
||||
min_detection_confidence=sensitivity
|
||||
) as detector:
|
||||
img_array = np.array(img)
|
||||
results = detector.process(img_array)
|
||||
|
||||
faces = []
|
||||
if results.detections:
|
||||
for detection in results.detections:
|
||||
bbox = detection.location_data.relative_bounding_box
|
||||
x = int(bbox.xmin * img.width)
|
||||
y = int(bbox.ymin * img.height)
|
||||
w = int(bbox.width * img.width)
|
||||
h = int(bbox.height * img.height)
|
||||
|
||||
# Add some padding around the face
|
||||
pad = int(max(w, h) * 0.1)
|
||||
x1 = max(0, x - pad)
|
||||
y1 = max(0, y - pad)
|
||||
x2 = min(img.width, x + w + pad)
|
||||
y2 = min(img.height, y + h + pad)
|
||||
|
||||
face_region = img.crop((x1, y1, x2, y2))
|
||||
blurred = face_region.filter(
|
||||
ImageFilter.GaussianBlur(blur_radius)
|
||||
)
|
||||
img.paste(blurred, (x1, y1))
|
||||
faces.append({"x": x, "y": y, "w": w, "h": h})
|
||||
|
||||
img.save(output_path)
|
||||
print(
|
||||
json.dumps(
|
||||
{
|
||||
"success": True,
|
||||
"facesDetected": len(faces),
|
||||
"faces": faces,
|
||||
}
|
||||
)
|
||||
)
|
||||
|
||||
except ImportError:
|
||||
# Fallback: no face detection available, save original
|
||||
img.save(output_path)
|
||||
print(
|
||||
json.dumps(
|
||||
{
|
||||
"success": True,
|
||||
"facesDetected": 0,
|
||||
"faces": [],
|
||||
"note": "mediapipe not available - no faces detected",
|
||||
}
|
||||
)
|
||||
)
|
||||
|
||||
except ImportError:
|
||||
print(
|
||||
json.dumps(
|
||||
{
|
||||
"success": False,
|
||||
"error": "Pillow is not installed. Install with: pip install Pillow",
|
||||
}
|
||||
)
|
||||
)
|
||||
sys.exit(1)
|
||||
except Exception as e:
|
||||
print(json.dumps({"success": False, "error": str(e)}))
|
||||
sys.exit(1)
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
main()
|
||||
@@ -0,0 +1,69 @@
|
||||
"""Object erasing / inpainting using LaMa or simple fallback."""
|
||||
import sys
|
||||
import json
|
||||
|
||||
|
||||
def main():
|
||||
input_path = sys.argv[1]
|
||||
mask_path = sys.argv[2]
|
||||
output_path = sys.argv[3]
|
||||
|
||||
try:
|
||||
from PIL import Image
|
||||
|
||||
try:
|
||||
# Try lama-cleaner if available
|
||||
from lama_cleaner.model_manager import ModelManager
|
||||
from lama_cleaner.schema import Config
|
||||
|
||||
img = Image.open(input_path).convert("RGB")
|
||||
mask = Image.open(mask_path).convert("L")
|
||||
|
||||
# Resize mask to match image if needed
|
||||
if mask.size != img.size:
|
||||
mask = mask.resize(img.size, Image.NEAREST)
|
||||
|
||||
import numpy as np
|
||||
|
||||
img_array = np.array(img)
|
||||
mask_array = np.array(mask)
|
||||
|
||||
model_manager = ModelManager(name="lama", device="cpu")
|
||||
config = Config(
|
||||
ldm_steps=25,
|
||||
ldm_sampler="plms",
|
||||
hd_strategy="Original",
|
||||
hd_strategy_crop_margin=128,
|
||||
hd_strategy_crop_trigger_size=800,
|
||||
hd_strategy_resize_limit=800,
|
||||
)
|
||||
result = model_manager(img_array, mask_array, config)
|
||||
Image.fromarray(result).save(output_path)
|
||||
method = "lama"
|
||||
|
||||
except (ImportError, Exception):
|
||||
# Fallback: simple inpainting using PIL
|
||||
# Just copy the image (mask areas won't be processed without ML model)
|
||||
img = Image.open(input_path)
|
||||
img.save(output_path)
|
||||
method = "copy"
|
||||
|
||||
print(json.dumps({"success": True, "method": method}))
|
||||
|
||||
except ImportError:
|
||||
print(
|
||||
json.dumps(
|
||||
{
|
||||
"success": False,
|
||||
"error": "Pillow is not installed. Install with: pip install Pillow",
|
||||
}
|
||||
)
|
||||
)
|
||||
sys.exit(1)
|
||||
except Exception as e:
|
||||
print(json.dumps({"success": False, "error": str(e)}))
|
||||
sys.exit(1)
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
main()
|
||||
@@ -0,0 +1,81 @@
|
||||
"""Text extraction from images using Tesseract or PaddleOCR."""
|
||||
import sys
|
||||
import json
|
||||
|
||||
|
||||
def main():
|
||||
input_path = sys.argv[1]
|
||||
settings = json.loads(sys.argv[2]) if len(sys.argv) > 2 else {}
|
||||
|
||||
engine = settings.get("engine", "tesseract")
|
||||
language = settings.get("language", "en")
|
||||
|
||||
try:
|
||||
if engine == "paddleocr":
|
||||
try:
|
||||
from paddleocr import PaddleOCR
|
||||
|
||||
ocr = PaddleOCR(use_angle_cls=True, lang=language)
|
||||
result = ocr.ocr(input_path, cls=True)
|
||||
text = "\n".join(
|
||||
[
|
||||
line[1][0]
|
||||
for res in result
|
||||
if res
|
||||
for line in res
|
||||
if line and line[1]
|
||||
]
|
||||
)
|
||||
print(
|
||||
json.dumps({"success": True, "text": text, "engine": "paddleocr"})
|
||||
)
|
||||
except ImportError:
|
||||
print(
|
||||
json.dumps(
|
||||
{
|
||||
"success": False,
|
||||
"error": "PaddleOCR is not installed. Install with: pip install paddleocr paddlepaddle",
|
||||
}
|
||||
)
|
||||
)
|
||||
sys.exit(1)
|
||||
else:
|
||||
# Tesseract via subprocess
|
||||
import subprocess
|
||||
|
||||
lang_map = {"en": "eng", "de": "deu", "fr": "fra", "es": "spa", "zh": "chi_sim", "ja": "jpn", "ko": "kor"}
|
||||
tess_lang = lang_map.get(language, "eng")
|
||||
|
||||
try:
|
||||
result = subprocess.run(
|
||||
["tesseract", input_path, "stdout", "-l", tess_lang],
|
||||
capture_output=True,
|
||||
text=True,
|
||||
timeout=120,
|
||||
)
|
||||
text = result.stdout.strip()
|
||||
if result.returncode != 0 and not text:
|
||||
raise RuntimeError(result.stderr.strip() or "Tesseract failed")
|
||||
print(
|
||||
json.dumps(
|
||||
{"success": True, "text": text, "engine": "tesseract"}
|
||||
)
|
||||
)
|
||||
except FileNotFoundError:
|
||||
print(
|
||||
json.dumps(
|
||||
{
|
||||
"success": False,
|
||||
"error": "Tesseract is not installed. Install with: apt-get install tesseract-ocr",
|
||||
}
|
||||
)
|
||||
)
|
||||
sys.exit(1)
|
||||
|
||||
except Exception as e:
|
||||
print(json.dumps({"success": False, "error": str(e)}))
|
||||
sys.exit(1)
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
main()
|
||||
@@ -0,0 +1,47 @@
|
||||
"""Background removal using rembg."""
|
||||
import sys
|
||||
import json
|
||||
|
||||
|
||||
def main():
|
||||
input_path = sys.argv[1]
|
||||
output_path = sys.argv[2]
|
||||
settings = json.loads(sys.argv[3]) if len(sys.argv) > 3 else {}
|
||||
|
||||
model = settings.get("model", "u2net")
|
||||
|
||||
try:
|
||||
from rembg import remove
|
||||
|
||||
with open(input_path, "rb") as f:
|
||||
input_data = f.read()
|
||||
|
||||
output_data = remove(
|
||||
input_data,
|
||||
alpha_matting=True,
|
||||
alpha_matting_foreground_threshold=240,
|
||||
alpha_matting_background_threshold=10,
|
||||
)
|
||||
|
||||
with open(output_path, "wb") as f:
|
||||
f.write(output_data)
|
||||
|
||||
print(json.dumps({"success": True, "model": model}))
|
||||
|
||||
except ImportError:
|
||||
print(
|
||||
json.dumps(
|
||||
{
|
||||
"success": False,
|
||||
"error": "rembg is not installed. Install with: pip install rembg[cpu]",
|
||||
}
|
||||
)
|
||||
)
|
||||
sys.exit(1)
|
||||
except Exception as e:
|
||||
print(json.dumps({"success": False, "error": str(e)}))
|
||||
sys.exit(1)
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
main()
|
||||
@@ -0,0 +1,78 @@
|
||||
"""Image upscaling with Real-ESRGAN fallback to Lanczos."""
|
||||
import sys
|
||||
import json
|
||||
|
||||
|
||||
def main():
|
||||
input_path = sys.argv[1]
|
||||
output_path = sys.argv[2]
|
||||
settings = json.loads(sys.argv[3]) if len(sys.argv) > 3 else {}
|
||||
|
||||
scale = settings.get("scale", 2)
|
||||
|
||||
try:
|
||||
from PIL import Image
|
||||
|
||||
img = Image.open(input_path)
|
||||
new_size = (img.width * scale, img.height * scale)
|
||||
|
||||
# Try Real-ESRGAN first
|
||||
try:
|
||||
from basicsr.archs.rrdbnet_arch import RRDBNet
|
||||
from realesrgan import RealESRGANer
|
||||
import numpy as np
|
||||
|
||||
model = RRDBNet(
|
||||
num_in_ch=3,
|
||||
num_out_ch=3,
|
||||
num_feat=64,
|
||||
num_block=23,
|
||||
num_grow_ch=32,
|
||||
scale=scale,
|
||||
)
|
||||
upsampler = RealESRGANer(
|
||||
scale=scale,
|
||||
model_path=None,
|
||||
model=model,
|
||||
half=False,
|
||||
)
|
||||
img_array = np.array(img.convert("RGB"))
|
||||
output, _ = upsampler.enhance(img_array, outscale=scale)
|
||||
result = Image.fromarray(output)
|
||||
result.save(output_path)
|
||||
method = "realesrgan"
|
||||
except (ImportError, Exception):
|
||||
# Fallback to Lanczos upscaling
|
||||
img_upscaled = img.resize(new_size, Image.LANCZOS)
|
||||
img_upscaled.save(output_path)
|
||||
method = "lanczos"
|
||||
|
||||
print(
|
||||
json.dumps(
|
||||
{
|
||||
"success": True,
|
||||
"scale": scale,
|
||||
"width": new_size[0],
|
||||
"height": new_size[1],
|
||||
"method": method,
|
||||
}
|
||||
)
|
||||
)
|
||||
|
||||
except ImportError:
|
||||
print(
|
||||
json.dumps(
|
||||
{
|
||||
"success": False,
|
||||
"error": "Pillow is not installed. Install with: pip install Pillow",
|
||||
}
|
||||
)
|
||||
)
|
||||
sys.exit(1)
|
||||
except Exception as e:
|
||||
print(json.dumps({"success": False, "error": str(e)}))
|
||||
sys.exit(1)
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
main()
|
||||
@@ -0,0 +1,30 @@
|
||||
import { runPythonScript } from "./bridge.js";
|
||||
import { writeFile, readFile } from "node:fs/promises";
|
||||
import { join } from "node:path";
|
||||
|
||||
export interface RemoveBackgroundOptions {
|
||||
model?: "u2net" | "isnet";
|
||||
}
|
||||
|
||||
export async function removeBackground(
|
||||
inputBuffer: Buffer,
|
||||
outputDir: string,
|
||||
options: RemoveBackgroundOptions = {},
|
||||
): Promise<Buffer> {
|
||||
const inputPath = join(outputDir, "input_bg.png");
|
||||
const outputPath = join(outputDir, "output_bg.png");
|
||||
|
||||
await writeFile(inputPath, inputBuffer);
|
||||
const { stdout } = await runPythonScript("remove_bg.py", [
|
||||
inputPath,
|
||||
outputPath,
|
||||
JSON.stringify(options),
|
||||
]);
|
||||
|
||||
const result = JSON.parse(stdout);
|
||||
if (!result.success) {
|
||||
throw new Error(result.error || "Background removal failed");
|
||||
}
|
||||
|
||||
return readFile(outputPath);
|
||||
}
|
||||
@@ -0,0 +1,86 @@
|
||||
import { execFile } from "node:child_process";
|
||||
import { promisify } from "node:util";
|
||||
import { resolve, dirname } from "node:path";
|
||||
import { fileURLToPath } from "node:url";
|
||||
|
||||
const execFileAsync = promisify(execFile);
|
||||
|
||||
const __dirname = dirname(fileURLToPath(import.meta.url));
|
||||
const PYTHON_DIR = resolve(__dirname, "../python");
|
||||
|
||||
/** Try venv first, then system python. */
|
||||
function getPythonPath(): string {
|
||||
const venvPath = process.env.PYTHON_VENV_PATH || "/opt/venv";
|
||||
return `${venvPath}/bin/python3`;
|
||||
}
|
||||
|
||||
/**
|
||||
* Extract a user-friendly error from a Python process error.
|
||||
* Python scripts print JSON to stderr/stdout on failure — try to parse it.
|
||||
*/
|
||||
function extractPythonError(error: unknown): string {
|
||||
if (error && typeof error === "object") {
|
||||
const execError = error as {
|
||||
stderr?: string;
|
||||
stdout?: string;
|
||||
message?: string;
|
||||
};
|
||||
// Try to parse JSON error from stderr or stdout
|
||||
for (const output of [execError.stderr, execError.stdout]) {
|
||||
if (output) {
|
||||
try {
|
||||
const parsed = JSON.parse(output.trim());
|
||||
if (parsed.error) return parsed.error;
|
||||
} catch {
|
||||
// Not JSON, check for human-readable content
|
||||
const trimmed = output.trim();
|
||||
if (trimmed && !trimmed.startsWith("Traceback")) {
|
||||
return trimmed;
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
if (execError.message) return execError.message;
|
||||
}
|
||||
return String(error);
|
||||
}
|
||||
|
||||
/**
|
||||
* Run a Python script from packages/ai/python/ with the given arguments.
|
||||
* Falls back to system python3 if the venv is not available.
|
||||
*/
|
||||
export async function runPythonScript(
|
||||
scriptName: string,
|
||||
args: string[],
|
||||
timeoutMs = 300000, // 5 min default
|
||||
): Promise<{ stdout: string; stderr: string }> {
|
||||
const scriptPath = resolve(PYTHON_DIR, scriptName);
|
||||
const pythonPath = getPythonPath();
|
||||
|
||||
const execOpts = {
|
||||
timeout: timeoutMs,
|
||||
maxBuffer: 50 * 1024 * 1024, // 50MB
|
||||
};
|
||||
|
||||
try {
|
||||
const { stdout, stderr } = await execFileAsync(
|
||||
pythonPath,
|
||||
[scriptPath, ...args],
|
||||
execOpts,
|
||||
);
|
||||
return { stdout: stdout.trim(), stderr: stderr.trim() };
|
||||
} catch {
|
||||
// Try system python as fallback
|
||||
try {
|
||||
const { stdout, stderr } = await execFileAsync(
|
||||
"python3",
|
||||
[scriptPath, ...args],
|
||||
execOpts,
|
||||
);
|
||||
return { stdout: stdout.trim(), stderr: stderr.trim() };
|
||||
} catch (fallbackError: unknown) {
|
||||
const message = extractPythonError(fallbackError);
|
||||
throw new Error(message);
|
||||
}
|
||||
}
|
||||
}
|
||||
@@ -0,0 +1,49 @@
|
||||
import { runPythonScript } from "./bridge.js";
|
||||
import { writeFile, readFile } from "node:fs/promises";
|
||||
import { join } from "node:path";
|
||||
|
||||
export interface BlurFacesOptions {
|
||||
blurRadius?: number;
|
||||
sensitivity?: number;
|
||||
}
|
||||
|
||||
export interface FaceRegion {
|
||||
x: number;
|
||||
y: number;
|
||||
w: number;
|
||||
h: number;
|
||||
}
|
||||
|
||||
export interface BlurFacesResult {
|
||||
buffer: Buffer;
|
||||
facesDetected: number;
|
||||
faces: FaceRegion[];
|
||||
}
|
||||
|
||||
export async function blurFaces(
|
||||
inputBuffer: Buffer,
|
||||
outputDir: string,
|
||||
options: BlurFacesOptions = {},
|
||||
): Promise<BlurFacesResult> {
|
||||
const inputPath = join(outputDir, "input_faces.png");
|
||||
const outputPath = join(outputDir, "output_faces.png");
|
||||
|
||||
await writeFile(inputPath, inputBuffer);
|
||||
const { stdout } = await runPythonScript("detect_faces.py", [
|
||||
inputPath,
|
||||
outputPath,
|
||||
JSON.stringify(options),
|
||||
]);
|
||||
|
||||
const result = JSON.parse(stdout);
|
||||
if (!result.success) {
|
||||
throw new Error(result.error || "Face detection failed");
|
||||
}
|
||||
|
||||
const buffer = await readFile(outputPath);
|
||||
return {
|
||||
buffer,
|
||||
facesDetected: result.facesDetected,
|
||||
faces: result.faces ?? [],
|
||||
};
|
||||
}
|
||||
@@ -1 +1,18 @@
|
||||
export const AI_VERSION = "0.0.1";
|
||||
|
||||
export { runPythonScript } from "./bridge.js";
|
||||
export { removeBackground } from "./background-removal.js";
|
||||
export type { RemoveBackgroundOptions } from "./background-removal.js";
|
||||
export { upscale } from "./upscaling.js";
|
||||
export type { UpscaleOptions, UpscaleResult } from "./upscaling.js";
|
||||
export { extractText } from "./ocr.js";
|
||||
export type { OcrOptions, OcrResult } from "./ocr.js";
|
||||
export { blurFaces } from "./face-detection.js";
|
||||
export type {
|
||||
BlurFacesOptions,
|
||||
BlurFacesResult,
|
||||
FaceRegion,
|
||||
} from "./face-detection.js";
|
||||
export { inpaint } from "./inpainting.js";
|
||||
export { smartCrop } from "./smart-crop.js";
|
||||
export type { SmartCropOptions } from "./smart-crop.js";
|
||||
|
||||
@@ -0,0 +1,29 @@
|
||||
import { runPythonScript } from "./bridge.js";
|
||||
import { writeFile, readFile } from "node:fs/promises";
|
||||
import { join } from "node:path";
|
||||
|
||||
export async function inpaint(
|
||||
inputBuffer: Buffer,
|
||||
maskBuffer: Buffer,
|
||||
outputDir: string,
|
||||
): Promise<Buffer> {
|
||||
const inputPath = join(outputDir, "input_inpaint.png");
|
||||
const maskPath = join(outputDir, "mask_inpaint.png");
|
||||
const outputPath = join(outputDir, "output_inpaint.png");
|
||||
|
||||
await writeFile(inputPath, inputBuffer);
|
||||
await writeFile(maskPath, maskBuffer);
|
||||
|
||||
const { stdout } = await runPythonScript("inpaint.py", [
|
||||
inputPath,
|
||||
maskPath,
|
||||
outputPath,
|
||||
]);
|
||||
|
||||
const result = JSON.parse(stdout);
|
||||
if (!result.success) {
|
||||
throw new Error(result.error || "Inpainting failed");
|
||||
}
|
||||
|
||||
return readFile(outputPath);
|
||||
}
|
||||
@@ -0,0 +1,37 @@
|
||||
import { runPythonScript } from "./bridge.js";
|
||||
import { writeFile } from "node:fs/promises";
|
||||
import { join } from "node:path";
|
||||
|
||||
export interface OcrOptions {
|
||||
engine?: "tesseract" | "paddleocr";
|
||||
language?: string;
|
||||
}
|
||||
|
||||
export interface OcrResult {
|
||||
text: string;
|
||||
engine: string;
|
||||
}
|
||||
|
||||
export async function extractText(
|
||||
inputBuffer: Buffer,
|
||||
outputDir: string,
|
||||
options: OcrOptions = {},
|
||||
): Promise<OcrResult> {
|
||||
const inputPath = join(outputDir, "input_ocr.png");
|
||||
|
||||
await writeFile(inputPath, inputBuffer);
|
||||
const { stdout } = await runPythonScript("ocr.py", [
|
||||
inputPath,
|
||||
JSON.stringify(options),
|
||||
]);
|
||||
|
||||
const result = JSON.parse(stdout);
|
||||
if (!result.success) {
|
||||
throw new Error(result.error || "OCR failed");
|
||||
}
|
||||
|
||||
return {
|
||||
text: result.text,
|
||||
engine: result.engine,
|
||||
};
|
||||
}
|
||||
@@ -0,0 +1,22 @@
|
||||
import sharp from "sharp";
|
||||
|
||||
export interface SmartCropOptions {
|
||||
width: number;
|
||||
height: number;
|
||||
}
|
||||
|
||||
/**
|
||||
* Smart crop using Sharp's entropy-based attention cropping.
|
||||
* No Python needed — uses Sharp's built-in saliency detection.
|
||||
*/
|
||||
export async function smartCrop(
|
||||
inputBuffer: Buffer,
|
||||
options: SmartCropOptions,
|
||||
): Promise<Buffer> {
|
||||
return sharp(inputBuffer)
|
||||
.resize(options.width, options.height, {
|
||||
fit: "cover",
|
||||
position: sharp.strategy.attention,
|
||||
})
|
||||
.toBuffer();
|
||||
}
|
||||
@@ -0,0 +1,43 @@
|
||||
import { runPythonScript } from "./bridge.js";
|
||||
import { writeFile, readFile } from "node:fs/promises";
|
||||
import { join } from "node:path";
|
||||
|
||||
export interface UpscaleOptions {
|
||||
scale?: number;
|
||||
}
|
||||
|
||||
export interface UpscaleResult {
|
||||
buffer: Buffer;
|
||||
width: number;
|
||||
height: number;
|
||||
method: string;
|
||||
}
|
||||
|
||||
export async function upscale(
|
||||
inputBuffer: Buffer,
|
||||
outputDir: string,
|
||||
options: UpscaleOptions = {},
|
||||
): Promise<UpscaleResult> {
|
||||
const inputPath = join(outputDir, "input_upscale.png");
|
||||
const outputPath = join(outputDir, "output_upscale.png");
|
||||
|
||||
await writeFile(inputPath, inputBuffer);
|
||||
const { stdout } = await runPythonScript("upscale.py", [
|
||||
inputPath,
|
||||
outputPath,
|
||||
JSON.stringify(options),
|
||||
]);
|
||||
|
||||
const result = JSON.parse(stdout);
|
||||
if (!result.success) {
|
||||
throw new Error(result.error || "Upscaling failed");
|
||||
}
|
||||
|
||||
const buffer = await readFile(outputPath);
|
||||
return {
|
||||
buffer,
|
||||
width: result.width,
|
||||
height: result.height,
|
||||
method: result.method ?? "unknown",
|
||||
};
|
||||
}
|
||||
Generated
+9
@@ -35,6 +35,9 @@ importers:
|
||||
'@fastify/swagger-ui':
|
||||
specifier: ^5.2.0
|
||||
version: 5.2.5
|
||||
'@stirling-image/ai':
|
||||
specifier: workspace:*
|
||||
version: link:../../packages/ai
|
||||
'@stirling-image/image-engine':
|
||||
specifier: workspace:*
|
||||
version: link:../../packages/image-engine
|
||||
@@ -160,7 +163,13 @@ importers:
|
||||
'@stirling-image/shared':
|
||||
specifier: workspace:*
|
||||
version: link:../shared
|
||||
sharp:
|
||||
specifier: ^0.33.0
|
||||
version: 0.33.5
|
||||
devDependencies:
|
||||
'@types/node':
|
||||
specifier: ^22.0.0
|
||||
version: 22.19.15
|
||||
typescript:
|
||||
specifier: ^5.7.0
|
||||
version: 5.9.3
|
||||
|
||||
Reference in New Issue
Block a user