In this paper, we propose several effective techniques to train deep convolution neural network(CNN) for hand pose estimation, which is a complex regression problem. We use repeating-ladder-type learning rate and hierarchical training based on physical property grouping to escape from poor solution and improve the final prediction accuracy. These two techniques can migrate to other regression problems which share similar properties with the hand pose estimation problem. We also give several image pre-processing methods to improve the stability. During the training process, we use high quality synthesized images to pre-train the networks and then use noisy initial images for fine tuning. We test on the NYU Hand Pose Dataset. Our method achieves the state-of-the-art result in the mi...[ Read more ]

In this paper, we propose several effective techniques to train deep convolution neural network(CNN) for hand pose estimation, which is a complex regression problem. We use repeating-ladder-type learning rate and hierarchical training based on physical property grouping to escape from poor solution and improve the final prediction accuracy. These two techniques can migrate to other regression problems which share similar properties with the hand pose estimation problem. We also give several image pre-processing methods to improve the stability. During the training process, we use high quality synthesized images to pre-train the networks and then use noisy initial images for fine tuning. We test on the NYU Hand Pose Dataset. Our method achieves the state-of-the-art result in the middle precision region.