Cross-Domain Detection via Graph-Induced Prototype Alignment

@article{Xu2020CrossDomainDV,
  title={Cross-Domain Detection via Graph-Induced Prototype Alignment},
  author={Minghao Xu and Hang Wang and Bingbing Ni and Qi Tian and Wenjun Zhang},
  journal={2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
  year={2020},
  pages={12352-12361}
}
  • Minghao XuHang Wang Wenjun Zhang
  • Published 28 March 2020
  • Computer Science
  • 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
Applying the knowledge of an object detector trained on a specific domain directly onto a new domain is risky, as the gap between two domains can severely degrade model's performance. Furthermore, since different instances commonly embody distinct modal information in object detection scenario, the feature alignment of source and target domain is hard to be realized. To mitigate these problems, we propose a Graph-induced Prototype Alignment (GPA) framework to seek for category-level domain… 

Cross-Domain Object Detection with Mean-Teacher Transformer

This work proposes an end-to-end cross-domain detection transformer based on the mean teacher knowledge transfer (MTKT), which transfers knowledge between domains via pseudo labels and consistently outperforms the SOTA baseline with a large margin on scene adaptation and synthetic to real adaptation and the performance of each individual category.

RPN Prototype Alignment For Domain Adaptive Object Detector

This work proposes a simple yet effective method suitable for RPN feature alignment to generate high-quality pseudo label of proposals in target domain using the filtered detection results with IoU, and shows the effectiveness of the method against previous state-of-the-art methods.

SCAN: Cross Domain Object Detection with Semantic Conditioned Adaptation

This work proposes a novel Semantic Conditioned AdaptatioN (SCAN) framework such that well-modeled unbiased semantics can support semantic conditioned adaptation for precise domain adaptive object detection.

SIGMA: Semantic-complete Graph Matching for Domain Adaptive Object Detection

Domain Adaptive Object Detection (DAOD) leverages a labeled domain to learn an object detector generalizing to a novel domain free of annotations. Recent advances align class-conditional

Informative and Consistent Correspondence Mining for Cross-Domain Weakly Supervised Object Detection

This work proposes to learn pixel-wise cross- domain correspondences for more precise knowledge transfer through a novel cross-domain co-attention scheme trained as region competition, and achieves consistent improvements over existing approaches by a considerable margin.

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

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 and exceeds the state-of-the-art performance on benchmark datasets.

Multi-Granularity Alignment Domain Adaptation for Object Detection

A unified multi-granularity alignment based object detection framework towards domain-invariant feature learning that leverages not only the instance discriminability in different categories but also the category consistency between two domains.

Sequential Instance Refinement for Cross-Domain Object Detection in Images

A reinforcement learning based method, coined as sequential instance refinement, where two agents are learned to progressively refine both source and target instances by taking sequential actions to remove both outlier target instances and low-relevance source instances step by step is proposed.

Category Dictionary Guided Unsupervised Domain Adaptation for Object Detection

A category dictionary guided (CDG) UDA model for cross-domain object detection is proposed, which learns category-specific dictionaries from the source domain to represent the candidate boxes in target domain and a residual weighted self-training paradigm is developed to implicitly align source and target domains for detection model training.

AFAN: Augmented Feature Alignment Network for Cross-Domain Object Detection

This work proposes a novel augmented feature alignment network (AFAN) which integrates intermediate domain image generation and domain-adversarial training into a unified framework and significantly outperforms the state-of-the-art methods on standard benchmarks for both similar and dissimilar domain adaptations.
...

References

SHOWING 1-10 OF 54 REFERENCES

Exploring Object Relation in Mean Teacher for Cross-Domain Detection

This work presents Mean Teacher with Object Relations (MTOR) that novelly remolds Mean Teacher under the backbone of Faster R-CNN by integrating the object relations into the measure of consistency cost between teacher and student modules.

GCAN: Graph Convolutional Adversarial Network for Unsupervised Domain Adaptation

This work proposes an end-to-end Graph Convolutional Adversarial Network (GCAN) for unsupervised domain adaptation by jointly modeling data structure, domain label, and class label in a unified deep framework.

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

This work builds on the recent state-of-the-art Faster R-CNN model, and design two domain adaptation components, on image level and instance level, to reduce the domain discrepancy, based on $$-divergence theory.

Progressive Feature Alignment for Unsupervised Domain Adaptation

The Progressive Feature Alignment Network (PFAN) is proposed to align the discriminative features across domains progressively and effectively, via exploiting the intra-class variation in the target domain.

Adapting Object Detectors via Selective Cross-Domain Alignment

The key idea is to mine the discriminative regions, namely those that are directly pertinent to object detection, and focus on aligning them across both domains, and perform remarkably better than existing methods.

Learning Semantic Representations for Unsupervised Domain Adaptation

Moving semantic transfer network is presented, which learn semantic representations for unlabeled target samples by aligning labeled source centroid and pseudo-labeled target centroid, resulting in an improved target classification accuracy.

Strong-Weak Distribution Alignment for Adaptive Object Detection

This work proposes an approach for unsupervised adaptation of object detectors from label-rich to label-poor domains which can significantly reduce annotation costs associated with detection, and designs the strong domain alignment model to only look at local receptive fields of the feature map.

Learning Context Graph for Person Search

This work proposes a contextual instance expansion module, which employs a relative attention module to search and filter useful context information in the scene, and builds a graph learning framework to effectively employ context pairs to update target similarity.

Diversify and Match: A Domain Adaptive Representation Learning Paradigm for Object Detection

This work introduces a novel unsupervised domain adaptation approach for object detection that outperforms the state-of-the-art methods by a large margin in terms of mean average precision (mAP) on various datasets.

A Robust Learning Approach to Domain Adaptive Object Detection

A robust object detection framework that is resilient to noise in bounding box class labels, locations and size annotations is proposed that significantly improves the state-of-the-art on multiple domain adaptation scenarios on the SIM10K, Cityscapes and KITTI datasets.
...