Exploring Object Relation in Mean Teacher for Cross-Domain Detection
arXiv:1904.11245
Abstract
Rendering synthetic data (e.g., 3D CAD-rendered images) to generate annotations for learning deep models in vision tasks has attracted increasing attention in recent years. However, simply applying the models learnt on synthetic images may lead to high generalization error on real images due to domain shift. To address this issue, recent progress in cross-domain recognition has featured the Mean Teacher, which directly simulates unsupervised domain adaptation as semi-supervised learning. The domain gap is thus naturally bridged with consistency regularization in a teacher-student scheme. In this work, we advance this Mean Teacher paradigm to be applicable for cross-domain detection. Specifically, we present 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. Technically, MTOR firstly learns relational graphs that capture similarities between pairs of regions for teacher and student respectively. The whole architecture is then optimized with three consistency regularizations: 1) region-level consistency to align the region-level predictions between teacher and student, 2) inter-graph consistency for matching the graph structures between teacher and student, and 3) intra-graph consistency to enhance the similarity between regions of same class within the graph of student. Extensive experiments are conducted on the transfers across Cityscapes, Foggy Cityscapes, and SIM10k, and superior results are reported when comparing to state-of-the-art approaches. More remarkably, we obtain a new record of single model: 22.8% of mAP on Syn2Real detection dataset.
arXiv commentCVPR 2019; The codes and model of our MTOR are publicly available at: https://github.com/caiqi/mean-teacher-cross-domain-detection
References in corpus (17)
- Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
- YOLOv3: An Incremental Improvement
- R-FCN: Object Detection via Region-based Fully Convolutional Networks
- Learning Transferable Features with Deep Adaptation Networks
- MXNet: A Flexible and Efficient Machine Learning Library for Heterogeneous Distributed Systems
- DSSD : Deconvolutional Single Shot Detector
- Mean teachers are better role models: Weight-averaged consistency targets improve semi-supervised deep learning results
- Semantic Foggy Scene Understanding with Synthetic Data
- Unsupervised Domain Adaptation with Residual Transfer Networks
- The Cityscapes Dataset for Semantic Urban Scene Understanding
- FCNs in the Wild: Pixel-level Adversarial and Constraint-based Adaptation
- Adversarial Discriminative Domain Adaptation
- Asymmetric Tri-training for Unsupervised Domain Adaptation
- Light-Head R-CNN: In Defense of Two-Stage Object Detector
- Fully Convolutional Adaptation Networks for Semantic Segmentation
- Syn2Real: A New Benchmark forSynthetic-to-Real Visual Domain Adaptation
- YouTube-BoundingBoxes: A Large High-Precision Human-Annotated Data Set for Object Detection in Video