Differentiable Multi-Granularity Human Representation Learning for Instance-Aware Human Semantic Parsing

TL;DR

This paper introduces a novel end-to-end framework for instance-aware human parsing that combines semantic segmentation and pose estimation using a differentiable grouping method, improving accuracy and efficiency.

Contribution

It proposes a multi-granularity learning approach with a differentiable bipartite matching for pixel grouping, enabling direct supervision and eliminating complex post-processing.

Findings

Outperforms existing bottom-up methods in accuracy.

Achieves more efficient inference on multiple datasets.

Enables end-to-end training with direct error back-propagation.

Abstract

To address the challenging task of instance-aware human part parsing, a new bottom-up regime is proposed to learn category-level human semantic segmentation as well as multi-person pose estimation in a joint and end-to-end manner. It is a compact, efficient and powerful framework that exploits structural information over different human granularities and eases the difficulty of person partitioning. Specifically, a dense-to-sparse projection field, which allows explicitly associating dense human semantics with sparse keypoints, is learnt and progressively improved over the network feature pyramid for robustness. Then, the difficult pixel grouping problem is cast as an easier, multi-person joint assembling task. By formulating joint association as maximum-weight bipartite matching, a differentiable solution is developed to exploit projected gradient descent and Dykstra's cyclic projection algorithm. This makes our method end-to-end trainable and allows back-propagating the grouping error to directly supervise multi-granularity human representation learning. This is distinguished from current bottom-up human parsers or pose estimators which require sophisticated post-processing or heuristic greedy algorithms. Experiments on three instance-aware human parsing datasets show that our model outperforms other bottom-up alternatives with much more efficient inference.