Overview

We offer a benchmark suite together with an evaluation server, such that authors can upload their results and get a ranking. We offer a Dataset that contains 38280 pictures, including 28280 images for training set, 5000 images for validation set and 5000 images for test set. If you would like to submit your results, please register, login, and follow the instructions on our submission page.

Note: We only display the highest submission of each person.

Multi-Person Human Parsing Task

Metrics

For this task, we use two metrics for multi-human parsing evaluation. Mean IoU(%) for semantic part segmentation, reported by the FCN paper. And AP^r for instance-level human parsing, first introduced in Simultaneous Detection and Segmentation. The details show AP^r at various IoU thresholds.