Pseudo Supervised Monocular Depth Estimation with Teacher-Student Network
This addresses a major hurdle in depth estimation for applications like autonomous driving, though it is incremental as it builds on existing unsupervised and supervised techniques.
The paper tackles the lack of high-quality ground truth annotations in monocular depth estimation by proposing an unsupervised method using a teacher-student network with knowledge distillation, which outperforms state-of-the-art on the KITTI benchmark.
Despite recent improvement of supervised monocular depth estimation, the lack of high quality pixel-wise ground truth annotations has become a major hurdle for further progress. In this work, we propose a new unsupervised depth estimation method based on pseudo supervision mechanism by training a teacher-student network with knowledge distillation. It strategically integrates the advantages of supervised and unsupervised monocular depth estimation, as well as unsupervised binocular depth estimation. Specifically, the teacher network takes advantage of the effectiveness of binocular depth estimation to produce accurate disparity maps, which are then used as the pseudo ground truth to train the student network for monocular depth estimation. This effectively converts the problem of unsupervised learning to supervised learning. Our extensive experimental results demonstrate that the proposed method outperforms the state-of-the-art on the KITTI benchmark.