I wanted to sharpen and upscale images without uploading them to some server. It turns out you can run the whole ML model right in the browser — the image never leaves the device. Here's what I learned shipping it as two free tools.
The stack
- TensorFlow.js (WebGL backend) for inference
- UpscalerJS as a thin wrapper around an ESRGAN super-resolution model
- All loaded from a CDN — no build step, no backend
The core is one call: new Upscaler({ model }), then await upscaler.upscale(img, { patchSize: 64, padding: 4, progress }). The patchSize option is the important one — more on that below.
Gotchas I hit
Memory. Running 4× on a large image tries to allocate a huge tensor and the tab dies. The fix is patchSize — process the image in tiles and stitch them back, so memory stays bounded regardless of input size.
Model choice matters more than I expected. I benchmarked three ESRGAN variants on the same image: slim is fast (~2.5s on a small image) but slightly soft, medium had visible tiling artifacts (rejected), and thick was clearly the sharpest but ~3× slower. So I default to slim and offer thick as a "max detail" mode.
Lightweight upscalers smooth the image. ESRGAN-slim enlarges cleanly but the result can look soft. A small unsharp-mask pass afterward restores the bite without an obvious "sharpened" halo.
Sharpen vs upscale are different jobs. To make a sharpener that keeps the original size, I run the same model then draw the result back down to native dimensions — the AI detail survives the downscale, so you get a clearer image at the same size. That became the AI photo sharpener; the enlarge-2×/4× version is the AI image upscaler.
Why keep it client-side?
- Privacy — the image is never uploaded. Great for documents, IDs, personal photos.
- Cost — it's a static site on Cloudflare Pages, so there's no inference server to pay for.
The trade-off is a one-time model download and slower runs on weak phones, which I gate with size caps and a fast/max toggle.
Happy to answer anything about the TF.js side — what would you run in-browser next?
Top comments (0)