ISNet (int8 weights, 44 MB) running on onnxruntime-web 1.22. The image is decoded, segmented and composited on your own device — nothing is uploaded.
Pick a photo, then run it with each execution provider and compare the timings. Try a portrait, a product shot and something with a busy background. The first inference is always slower (warm-up and shader compilation) — read the second one.
No file chosen
Waiting…
Production version: is.tools/remove-background · Source and benchmarks: GitHub