Exploring the Robustness of Human Parsers Towards Common Corruptions

TL;DR

This paper introduces new benchmarks and a novel augmentation-based method to significantly improve the robustness of human parsers against common image corruptions like blur and noise, without sacrificing performance on clean data.

Contribution

The paper proposes a heterogeneous augmentation-enhanced mechanism combining image-aware and model-aware augmentations to improve robustness of human parsers under corruptions, applicable to various models.

Findings

Improved robustness of human parsers on corruption benchmarks

Method enhances model resilience without losing clean data accuracy

Universal applicability across different human parsing frameworks

Abstract

Human parsing aims to segment each pixel of the human image with fine-grained semantic categories. However, current human parsers trained with clean data are easily confused by numerous image corruptions such as blur and noise. To improve the robustness of human parsers, in this paper, we construct three corruption robustness benchmarks, termed LIP-C, ATR-C, and Pascal-Person-Part-C, to assist us in evaluating the risk tolerance of human parsing models. Inspired by the data augmentation strategy, we propose a novel heterogeneous augmentation-enhanced mechanism to bolster robustness under commonly corrupted conditions. Specifically, two types of data augmentations from different views, i.e., image-aware augmentation and model-aware image-to-image transformation, are integrated in a sequential manner for adapting to unforeseen image corruptions. The image-aware augmentation can enrich the high diversity of training images with the help of common image operations. The model-aware augmentation strategy that improves the diversity of input data by considering the model's randomness. The proposed method is model-agnostic, and it can plug and play into arbitrary state-of-the-art human parsing frameworks. The experimental results show that the proposed method demonstrates good universality which can improve the robustness of the human parsing models and even the semantic segmentation models when facing various image common corruptions. Meanwhile, it can still obtain approximate performance on clean data.