Weakly-Supervised Learning of Category-specific 3D Object Shapes.

Category-specific 3D object shape models have greatly boosted the recent advances in object detection, recognition and segmentation. However, even the most advanced approach for learning 3D object shapes still requires heavy manual annotations on large-scale 2D images. In particular, annotating figure-ground segmentation is unbearably labor-intensive and time-consuming. To alleviate the costs of such manual annotations, we make an effort to learn category-specific 3D shape models by using weakly-labeled 2D images, where only object categories and keypoints are annotated. By exploring the underlying relationship between two tasks: object segmentation and category-specific 3D shape reconstruction, we propose a novel weakly-supervised learning framework to jointly address these two tasks and collaborate them to boost the final performance of the learned 3D shape models. Moreover, learning without using figure-ground segmentation leads to ambiguous solutions. To this end, we develop the confidence weighting schemes in the viewpoint estimation and 3D shape learning procedure. These schemes effectively reduce the confusion caused by the noisy data and thus increase the chances for obtaining more reliable 3D object shapes. Comprehensive experiments on the challenging PASCAL VOC benchmark show that our framework achieves comparable performance of the state-of-the-art methods that use expensive manual segmentation annotations.