PACF: Prototype Augmented Compact Features for Improving Domain Adaptive Object Detection

TL;DR

This paper introduces PACF, a novel framework that enhances domain adaptive object detection by regularizing feature distributions, reducing variance and mean shift, and achieving state-of-the-art results across various settings.

Contribution

The paper proposes a new prototype-based regularization method with theoretical analysis and mutual learning strategies to improve cross-domain feature compactness and discriminability.

Findings

Significantly reduced variance of target features' class-conditional distributions.

Decreased class-mean shift between source and target domains.

Achieved state-of-the-art performance in multiple domain adaptation settings.

Abstract

In recent years, there has been significant advancement in object detection. However, applying off-the-shelf detectors to a new domain leads to significant performance drop, caused by the domain gap. These detectors exhibit higher-variance class-conditional distributions in the target domain than that in the source domain, along with mean shift. To address this problem, we propose the Prototype Augmented Compact Features (PACF) framework to regularize the distribution of intra-class features. Specifically, we provide an in-depth theoretical analysis on the lower bound of the target features-related likelihood and derive the prototype cross entropy loss to further calibrate the distribution of target RoI features. Furthermore, a mutual regularization strategy is designed to enable the linear and prototype-based classifiers to learn from each other, promoting feature compactness while enhancing discriminability. Thanks to this PACF framework, we have obtained a more compact cross-domain feature space, within which the variance of the target features' class-conditional distributions has significantly decreased, and the class-mean shift between the two domains has also been further reduced. The results on different adaptation settings are state-of-the-art, which demonstrate the board applicability and effectiveness of the proposed approach.