Distilling Knowledge via Knowledge Review
This work addresses a bottleneck in knowledge distillation for improving student network performance across various computer vision tasks, though it appears incremental as it builds on existing methods by focusing on connection paths.
The paper tackles the problem of knowledge distillation by proposing cross-stage connection paths between teacher and student networks, revealing their importance and leading to significant performance improvements in tasks like classification, object detection, and instance segmentation.
Knowledge distillation transfers knowledge from the teacher network to the student one, with the goal of greatly improving the performance of the student network. Previous methods mostly focus on proposing feature transformation and loss functions between the same level's features to improve the effectiveness. We differently study the factor of connection path cross levels between teacher and student networks, and reveal its great importance. For the first time in knowledge distillation, cross-stage connection paths are proposed. Our new review mechanism is effective and structurally simple. Our finally designed nested and compact framework requires negligible computation overhead, and outperforms other methods on a variety of tasks. We apply our method to classification, object detection, and instance segmentation tasks. All of them witness significant student network performance improvement. Code is available at https://github.com/Jia-Research-Lab/ReviewKD