I3Net: Implicit Instance-Invariant Network for Adapting One-Stage Object Detectors

TL;DR

This paper introduces I3Net, a novel approach for adapting one-stage object detectors across domains by implicitly learning instance-invariant features through multi-faceted feature alignment and reweighting strategies.

Contribution

The paper proposes I3Net, which uniquely adapts one-stage detectors without explicit instance features, using dynamic reweighting, category-aware pattern matching, and joint category alignment.

Findings

I3Net outperforms existing methods on benchmark datasets.

The proposed modules improve cross-domain detection accuracy.

Implicit feature learning effectively replaces explicit instance features.

Abstract

Recent works on two-stage cross-domain detection have widely explored the local feature patterns to achieve more accurate adaptation results. These methods heavily rely on the region proposal mechanisms and ROI-based instance-level features to design fine-grained feature alignment modules with respect to the foreground objects. However, for one-stage detectors, it is hard or even impossible to obtain explicit instance-level features in the detection pipelines. Motivated by this, we propose an Implicit Instance-Invariant Network (I3Net), which is tailored for adapting one-stage detectors and implicitly learns instance-invariant features via exploiting the natural characteristics of deep features in different layers. Specifically, we facilitate the adaptation from three aspects: (1) Dynamic and Class-Balanced Reweighting (DCBR) strategy, which considers the coexistence of intra-domain and intra-class variations to assign larger weights to those sample-scarce categories and easy-to-adapt samples; (2) Category-aware Object Pattern Matching (COPM) module, which boosts the cross-domain foreground objects matching guided by the categorical information and suppresses the uninformative background features; (3) Regularized Joint Category Alignment (RJCA) module, which jointly enforces the category alignment at different domain-specific layers with a consistency regularization. Experiments reveal that I3Net exceeds the state-of-the-art performance on benchmark datasets.