Implicit Non-Causal Factors are Out via Dataset Splitting for Domain Generalization Object Detection

TL;DR

This paper introduces GB-DAL, a novel domain generalization method for object detection that leverages dataset splitting and data augmentation to better identify and utilize implicit non-causal factors, improving generalization.

Contribution

It proposes GB-DAL with PGBS and SNF modules to generate dense domains and simulate non-causal factors, enhancing domain-invariant representation learning.

Findings

Outperforms existing methods on multiple benchmarks.

Improves domain generalization in open world object detection.

Effectively captures implicit non-causal factors.

Abstract

Open world object detection faces a significant challenge in domain-invariant representation, i.e., implicit non-causal factors. Most domain generalization (DG) methods based on domain adversarial learning (DAL) pay much attention to learn domain-invariant information, but often overlook the potential non-causal factors. We unveil two critical causes: 1) The domain discriminator-based DAL method is subject to the extremely sparse domain label, i.e., assigning only one domain label to each dataset, thus can only associate explicit non-causal factor, which is incredibly limited. 2) The non-causal factors, induced by unidentified data bias, are excessively implicit and cannot be solely discerned by conventional DAL paradigm. Based on these key findings, inspired by the Granular-Ball perspective, we propose an improved DAL method, i.e., GB-DAL. The proposed GB-DAL utilizes Prototype-based Granular Ball Splitting (PGBS) module to generate more dense domains from limited datasets, akin to more fine-grained granular balls, indicating more potential non-causal factors. Inspired by adversarial perturbations akin to non-causal factors, we propose a Simulated Non-causal Factors (SNF) module as a means of data augmentation to reduce the implicitness of non-causal factors, and facilitate the training of GB-DAL. Comparative experiments on numerous benchmarks demonstrate that our method achieves better generalization performance in novel circumstances.