Unsupervised Domain Adaptation through Inter-modal Rotation for RGB-D Object Recognition
This work addresses the challenge of adapting multi-modal RGB-D data from synthetic to real domains in robotics, offering an incremental improvement over existing methods.
The paper tackles the problem of unsupervised domain adaptation for RGB-D object recognition by proposing a method that reduces synthetic-to-real domain shift through inter-modal rotation prediction, achieving improved performance on new benchmark datasets for object categorization and instance recognition.
Unsupervised Domain Adaptation (DA) exploits the supervision of a label-rich source dataset to make predictions on an unlabeled target dataset by aligning the two data distributions. In robotics, DA is used to take advantage of automatically generated synthetic data, that come with "free" annotation, to make effective predictions on real data. However, existing DA methods are not designed to cope with the multi-modal nature of RGB-D data, which are widely used in robotic vision. We propose a novel RGB-D DA method that reduces the synthetic-to-real domain shift by exploiting the inter-modal relation between the RGB and depth image. Our method consists of training a convolutional neural network to solve, in addition to the main recognition task, the pretext task of predicting the relative rotation between the RGB and depth image. To evaluate our method and encourage further research in this area, we define two benchmark datasets for object categorization and instance recognition. With extensive experiments, we show the benefits of leveraging the inter-modal relations for RGB-D DA.