Knowledge distillation facilitates the training of a compact student network by using a deep teacher one. While this has achieved great success in many tasks, it remains completely unstudied for image-based 6D object pose estimation. In this work, we introduce the first knowledge distillation method driven by the 6D pose estimation task. To this end, we observe that most modern 6D pose estimation frameworks output local predictions, such as sparse 2D keypoints or dense representations, and that the compact student network typically struggles to predict such local quantities precisely. Therefore, instead of imposing prediction-to-prediction supervision from the teacher to the student, we propose to distill the teacher's \emph{distribution} of local predictions into the student network, facilitating its training. Our experiments on several benchmarks show that our distillation method yields state-of-the-art results with different compact student models and for both keypoint-based and dense prediction-based architectures.
翻译:知识蒸馏有助于通过使用深层教师来培训紧凑的学生网络。 虽然它在许多任务中取得了巨大成功, 但对于基于图像的 6D 对象来说, 仍然完全没有被研究出来。 在这项工作中, 我们引入了由 6D 构成的估算任务驱动的第一种知识蒸馏方法。 为此, 我们观察到, 大多数现代的 6D 构成估算框架, 输出本地预测, 如稀疏的 2D 关键点或密度表示, 并且 紧凑的学生网络通常很难准确预测本地数量 。 因此, 我们不把教师对本地预测的预测- 预测监督强加给学生, 而是建议将教师对本地预测的 \ emph{ 分布 推荐 推荐 提炼到学生网络中, 方便学生网络的培训。 我们在几个基准上的实验显示, 我们的蒸馏方法以不同的紧凑学生模型和基于关键点和密集预测的架构, 产生最新的最新结果 。