Does this tool support batch processing for corners, border, and shadow?
No. This tool is designed for single-image touch-ups, since the right corner radius or shadow strength usually depends on the individual image's proportions and purpose — for example, a square screenshot and a long banner need very different corner ratios, and applying the same settings to both often looks off. So it's designed for fine-tuning one image at a time. If you need to process multiple images, open and adjust them one by one.
Is this done on my computer or phone, or does it get uploaded to a server?
It's done entirely in your current device's browser. The live preview uses the browser's native CSS border, corner, and shadow styles for instant rendering, and when you download, the final image is redrawn with Canvas using the same parameters — your original image is never uploaded to any server.
After adding a shadow, why does the exported image end up larger than the original?
A shadow needs to extend outward some distance to be visible — if the canvas stayed the same size as the original, the shadow would get clipped and disappear. So on export, the canvas automatically expands based on your current shadow blur, offset distance, and border width, and the extra area is filled with your chosen background color (or kept transparent), ensuring the shadow displays fully without being cut off.
I set the background color to “transparent,” but the exported JPG still has a white background — why?
JPG doesn't support a transparency channel. If you choose a transparent background but the original is a JPG, the transparent area automatically becomes white — this is a fundamental limitation of the JPG format, not a processing error. If you need a truly transparent background, upload a PNG or WebP image, and the transparency will be preserved in the result.
Are the corner radius, border width, and shadow range fixed pixel values, or calculated as a proportion?
They're all calculated as a proportion of the image's own width and height, not fixed pixel values — so no matter whether the uploaded image is high or low resolution, the same slider value produces a visually consistent proportional effect. What you see in the preview matches exactly what you get in the downloaded final image.
Will the processed image keep its EXIF capture data or original transparency?
Processing fundamentally redraws and re-encodes the image: if the original has a transparent background, the transparent parts outside the rounded corners are preserved (unless you chose an opaque background color to fill it). The original's EXIF capture data and location metadata are not preserved after re-encoding — use the “EXIF Viewer/Remover” tool if you specifically need to view or remove that data.