Domain Adaptation for Object Detection via Style Consistency

Abstract

We propose a domain adaptation approach for object detection. We introduce a two-step method: the first step makes the detector robust to low-level differences and the second step adapts the classifiers to changes in the high-level features. For the first step, we use a style transfer method for pixel-adaptation of source images to the target domain. We find that enforcing low distance in the high-level features of the object detector between the style transferred images and the source images improves the performance in the target domain. For the second step, we propose a robust pseudo labelling approach to reduce the noise in both positive and negative sampling. Experimental evaluation is performed using the detector SSD300 on PASCAL VOC extended with the dataset proposed in arxiv:1803.11365 where the target domain images are of different styles. Our approach significantly improves the state-of-the-art performance in this benchmark.

BMVC 2019

References in corpus (12)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
• Semantic Foggy Scene Understanding with Synthetic Data
• A Learned Representation For Artistic Style
• ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustness
• Texture Networks: Feed-forward Synthesis of Textures and Stylized Images
• Weakly Supervised Object Localization with Multi-fold Multiple Instance Learning
• Asymmetric Tri-training for Unsupervised Domain Adaptation
• LSDA: Large Scale Detection Through Adaptation
• Photorealistic Style Transfer via Wavelet Transforms
• SPLAT: Semantic Pixel-Level Adaptation Transforms for Detection
• OMNIA Faster R-CNN: Detection in the wild through dataset merging and soft distillation

Cited by in corpus (8)

• From Handcrafted to Deep Features for Pedestrian Detection: A Survey
• Enhancing Human Pose Estimation in Ancient Vase Paintings via Perceptually-grounded Style Transfer Learning
• Cross-domain Object Detection through Coarse-to-Fine Feature Adaptation
• Deep Domain Adaptive Object Detection: a Survey
• SF-UDA${}^{3D}$: Source-Free Unsupervised Domain Adaptation for LiDAR-Based 3D Object Detection
• A Survey on Deep Domain Adaptation and Tiny Object Detection Challenges, Techniques and Datasets
• Exploiting Domain Transferability for Collaborative Inter-level Domain Adaptive Object Detection
• Memory-guided Unsupervised Image-to-image Translation