FootNet: A Multi-View Smartphone Dataset and Four-Model Benchmark for Clinical Foot Segmentation

We present FootNet, a 453-image multi-view smart-phone foot dataset for binary foot segmentation, with expert-annotated masks across six anatomical views (dorsal, medial, and plantar, both left and right). We benchmark four segmentation models under a controlled protocol: U-Net with a MobileNetV2 encoder achieves the best performance (IoU 0.9268, Dice 0.9608, 95% CI [0.9209, 0.9320]); DeepLabV3 with MobileNetV3-Large scores IoU 0.8984 (Dice 0.9449); UNet++ with MobileNetV2 scores IoU 0.8913 (Dice 0.9391); and SAM ViT-B with oracle bounding-box prompt scores IoU 0.9219 on the matched 191-image sub-set. Bonferroni-corrected Wilcoxon signed-rank tests ( k = 6 comparisons) show U-Net significantly outperforms DeepLab ( p < 0.001, r = 0.638) and SAM ViT-B with oracle bounding-box ( p = 0.005, r = 0.202); UNet++ does not significantly differ from DeepLab ( p = 0.062). Connected-component post-processing yields negligible benefit (mean ΔIoU = +0.0003, 12 of 453 images improved). The extended dataset is available upon request.