CIParsing: Unifying Causality Properties into Multiple Human Parsing

by   Xiaojia Chen, et al.

Existing methods of multiple human parsing (MHP) apply statistical models to acquire underlying associations between images and labeled body parts. However, acquired associations often contain many spurious correlations that degrade model generalization, leading statistical models to be vulnerable to visually contextual variations in images (e.g., unseen image styles/external interventions). To tackle this, we present a causality inspired parsing paradigm termed CIParsing, which follows fundamental causal principles involving two causal properties for human parsing (i.e., the causal diversity and the causal invariance). Specifically, we assume that an input image is constructed by a mix of causal factors (the characteristics of body parts) and non-causal factors (external contexts), where only the former ones cause the generation process of human parsing.Since causal/non-causal factors are unobservable, a human parser in proposed CIParsing is required to construct latent representations of causal factors and learns to enforce representations to satisfy the causal properties. In this way, the human parser is able to rely on causal factors w.r.t relevant evidence rather than non-causal factors w.r.t spurious correlations, thus alleviating model degradation and yielding improved parsing ability. Notably, the CIParsing is designed in a plug-and-play fashion and can be integrated into any existing MHP models. Extensive experiments conducted on two widely used benchmarks demonstrate the effectiveness and generalizability of our method.


page 4

page 8


Causality Inspired Representation Learning for Domain Generalization

Domain generalization (DG) is essentially an out-of-distribution problem...

Inducing Causal Structure for Abstractive Text Summarization

The mainstream of data-driven abstractive summarization models tends to ...

Generative Interventions for Causal Learning

We introduce a framework for learning robust visual representations that...

Towards Real World Human Parsing: Multiple-Human Parsing in the Wild

The recent progress of human parsing techniques has been largely driven ...

CDGNet: Class Distribution Guided Network for Human Parsing

The objective of human parsing is to partition a human in an image into ...

CITRIS: Causal Identifiability from Temporal Intervened Sequences

Understanding the latent causal factors of a dynamical system from visua...

KE-RCNN: Unifying Knowledge based Reasoning into Part-level Attribute Parsing

Part-level attribute parsing is a fundamental but challenging task, whic...

Please sign up or login with your details

Forgot password? Click here to reset