Abstract:Conventional object detectors often encounter remarkable performance drops due to the domain shift caused by environmental changes. However, labeling sufficient training data drawn from various domains is cost-ineffective and labor-intensive. To this end, unsupervised domain adaptive object detection (DAOD) has attracted much attention, in which the detector is transferred from the label-rich source domain to a label-agnostic target domain. Most of the existing cross-domain object detectors are designed with a parameter-shared network architecture, which, however, has an inherent flaw to accumulate the source errors caused by the inaccurate target distribution. As a result, the target data instead of decisive source data may dominate the learning towards model collapse. To overcome the risk, we propose an Asymmetric Tri-way Faster-RCNN (ATF) for DAOD tasks, in which a novel ancillary net is deployed with the chief net and it brings two merits. 1) The ancillary net enables the source distribution to be preserved without accumulating source error/distortion, such that the target distribution is dominated by source data and the model collapse is alleviated. 2) Ancillary target features can be generated by the ancillary net, which further enhances the discrimination of the target domain object detector. Furthermore, in order to remove the useless source-specific knowledge and exploit the informative domain-invariant knowledge, we propose a Partial Alignment based ATF (PA-ATF) model, in which an adversarial shuffling based mutual-information minimization strategy for domain disentanglement is provided. Intuitively, transferring only the source-domain invariant information to the target domain is more conscionable. Extensive experiments on benchmark datasets, including the Cityscapes, Foggy Cityscapes, Pascal VOC, Clipart, Watercolor, SIM10K, and KITTI demonstrate the remarkable performance of our models over other state-of-the-art approaches.

Domain Adaptation for Infrared-Radar Cross-Scene Multimodal Detection

Multiscale Attention-Based Subdomain Dynamic Adaptation for Cross-Domain Scene Classification

Enhanced Cross-Domain Dim and Small Infrared Target Detection via Content-Decoupled Feature Alignment

A Semantic Domain Adaption Framework for Cross-Domain Infrared Small Target Detection

Rotation-Invariant and Relation-Aware Cross-Domain Adaptation Object Detection Network for Optical Remote Sensing Images

Unsupervised Domain Adaptation for SAR Target Detection

Monocular 3D Object Detection via Feature Domain Adaptation

Cross-Scene Joint Classification of Multisource Data With Multilevel Domain Adaption Network

Infrared Small-Sample Target Recognition with Content-focused Domain Adaption Network

Open-Scenario Domain Adaptive Object Detection in Autonomous Driving

Partial Alignment for Object Detection in the Wild

Feature Aggregation and Compensation-Based Domain Adaptation Framework for Cross-Resolution Target Recognition in SAR Imagery

Two-Stage Domain Adaptation Based on Image and Feature Levels for Cloud Detection in Cross-Spatiotemporal Domain

Cross-Domain Object Detection by Dual Adaptive Branch

Multi-Level Domain Adaptive Learning For Cross-Domain Detection

Domain Adaptive Faster R-CNN for Object Detection in the Wild

Context-Guided Reverse Attention Network with Multiscale Aggregation for Infrared Small Target Detection

C3DA: A Universal Domain Adaptation Method for Scene Classification From Remote Sensing Imagery

Collaborative Training Between Region Proposal Localization and Classification for Domain Adaptive Object Detection

Domain Adaptation for Convolutional Neural Networks-Based Remote Sensing Scene Classification.

Dual-Domain Prior-Driven Deep Network for Infrared Small-Target Detection