Diversify and Match

  • Abstract

We propose a novel unsupervised domain adaptation method for object detection. We aim to alleviate limitations of feature-level and pixel-level domain adaptation approaches. Our approach is composed of various style translation and robust intra-class feature learning, and we construct a structured domain adaptation framework. Our method outperforms the state-of-the-art methods by a large margin in terms of mean average precision (mAP) on cartoon datasets.

  • Domain adaptation framework

Our proposed domain adaptation framework is composed of various style translation module, object detection module, and domain discriminator module. Various style translation module perturbs the input image to arbitrary cartoon styles. The translated source image is utilized to learn the large intra-class variance through domain discriminator. This scheme encourages the network to generate less domain-specific and more semantic features. Finally, these features are used for effective object detection on the target domain.

  • Various cartoon style transfer

We observe that varying the learning trend with alternative constraints causes the image translator to perturb the appearance of the translated images. Thus, we apply several variants of constraints to achieve distinct cartoon styles.

  • Quantitative results for the object detection of the cartoon style test set

We compared our method with source only and oracle methods on Faster R-CNN backbone. Our learning method achieved the higher class-wise AP than source only cases. Furthermore, we achieved the higher mean AP than oracle method, the supervision case with target (cartoon) domain labels.

  • Qualitative results for the object detection of the cartoon style test set

Taekyung Kim, Minki Jeong, Seunghyeon Kim, Seokeon Choi, and Changick Kim, "Diversify and Match: A Domain Adaptive Representation Learning Paradigm for Object Detection," Accepted to IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Long Beach, CA, USA, Jun. 16-20, 2019.