Harmonizing Transferability and Discriminability for Adapting Object Detectors

Abstract

Recent advances in adaptive object detection have achieved compelling results in virtue of adversarial feature adaptation to mitigate the distributional shifts along the detection pipeline. Whilst adversarial adaptation significantly enhances the transferability of feature representations, the feature discriminability of object detectors remains less investigated. Moreover, transferability and discriminability may come at a contradiction in adversarial adaptation given the complex combinations of objects and the differentiated scene layouts between domains. In this paper, we propose a Hierarchical Transferability Calibration Network (HTCN) that hierarchically (local-region/image/instance) calibrates the transferability of feature representations for harmonizing transferability and discriminability. The proposed model consists of three components: (1) Importance Weighted Adversarial Training with input Interpolation (IWAT-I), which strengthens the global discriminability by re-weighting the interpolated image-level features; (2) Context-aware Instance-Level Alignment (CILA) module, which enhances the local discriminability by capturing the underlying complementary effect between the instance-level feature and the global context information for the instance-level feature alignment; (3) local feature masks that calibrate the local transferability to provide semantic guidance for the following discriminative pattern alignment. Experimental results show that HTCN significantly outperforms the state-of-the-art methods on benchmark datasets.

arXiv commentAccepted by CVPR 2020

References in corpus (15)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
• How transferable are features in deep neural networks?
• Learning Transferable Features with Deep Adaptation Networks
• Unsupervised Domain Adaptation by Backpropagation
• Deep Domain Confusion: Maximizing for Domain Invariance
• Conditional Adversarial Domain Adaptation
• Adversarial Discriminative Domain Adaptation
• A DIRT-T Approach to Unsupervised Domain Adaptation
• Wasserstein Distance Guided Representation Learning for Domain Adaptation
• Random Feature Maps for Dot Product Kernels
• Co-regularized Alignment for Unsupervised Domain Adaptation
• Contrastive Adaptation Network for Unsupervised Domain Adaptation
• Transferrable Prototypical Networks for Unsupervised Domain Adaptation
• Penalizing Top Performers: Conservative Loss for Semantic Segmentation Adaptation

Cited by in corpus (3)

• Relation Matters: Foreground-aware Graph-based Relational Reasoning for Domain Adaptive Object Detection
• Rethinking Cross-Domain Pedestrian Detection: A Background-Focused Distribution Alignment Framework for Instance-Free One-Stage Detectors
• Exploiting Domain Transferability for Collaborative Inter-level Domain Adaptive Object Detection