Learning Dense Correspondence via 3D-guided Cycle Consistency
This addresses the challenge of learning correspondences for tasks where labels are difficult to obtain, such as in computer vision, but it is incremental as it builds on existing cycle-consistency ideas with a 3D twist.
The paper tackles the problem of establishing dense visual correspondence across different object instances without human-labeled ground truth by using 3D-guided cycle consistency as a supervisory signal, resulting in a ConvNet that outperforms state-of-the-art pairwise matching methods in correspondence-related tasks.
Discriminative deep learning approaches have shown impressive results for problems where human-labeled ground truth is plentiful, but what about tasks where labels are difficult or impossible to obtain? This paper tackles one such problem: establishing dense visual correspondence across different object instances. For this task, although we do not know what the ground-truth is, we know it should be consistent across instances of that category. We exploit this consistency as a supervisory signal to train a convolutional neural network to predict cross-instance correspondences between pairs of images depicting objects of the same category. For each pair of training images we find an appropriate 3D CAD model and render two synthetic views to link in with the pair, establishing a correspondence flow 4-cycle. We use ground-truth synthetic-to-synthetic correspondences, provided by the rendering engine, to train a ConvNet to predict synthetic-to-real, real-to-real and real-to-synthetic correspondences that are cycle-consistent with the ground-truth. At test time, no CAD models are required. We demonstrate that our end-to-end trained ConvNet supervised by cycle-consistency outperforms state-of-the-art pairwise matching methods in correspondence-related tasks.