Human Segmentation Tool

Upload an image or a video, select a model and a task. Several segmentation pipelines are available — each with different strengths depending on your input and use case.

Note: Running on CPU — inference takes a few seconds per image, and video processing time depends on video length and model choice.

Select Test Type
Select Model

Two-stage pipeline. YOLO26 locates each person with a bounding box, then UNET (ResNet-34 encoder) segments the body within each crop. Fast and instance-aware — handles multiple people separately.

Select Task

Original image with detected person highlighted in green. Useful for inspecting model output quality.

Example Images

Model: YOLO26s (detection) + UNET ResNet-50 (segmentation)
Training data: LIP · COCO-person · MADS · Penn-Fudan
GitHub: Human Segmentation Evaluation