Efficient Object Localization Using Convolutional Networks
arXiv:1411.4280
Abstract
Recent state-of-the-art performance on human-body pose estimation has been achieved with Deep Convolutional Networks (ConvNets). Traditional ConvNet architectures include pooling and sub-sampling layers which reduce computational requirements, introduce invariance and prevent over-training. These benefits of pooling come at the cost of reduced localization accuracy. We introduce a novel architecture which includes an efficient `position refinement' model that is trained to estimate the joint offset location within a small region of the image. This refinement model is jointly trained in cascade with a state-of-the-art ConvNet model to achieve improved accuracy in human joint location estimation. We show that the variance of our detector approaches the variance of human annotations on the FLIC dataset and outperforms all existing approaches on the MPII-human-pose dataset.
8 pages with 1 page of citations
References in corpus (3)
Cited by in corpus (11)
- Recent Advances in Convolutional Neural Networks
- Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields
- Deep Learning-based Extreme Heatwave Forecast
- Multi-Sample Dropout for Accelerated Training and Better Generalization
- Unconstrained Facial Landmark Localization with Backbone-Branches Fully-Convolutional Networks
- Generalised gravitational burst generation with Generative Adversarial Networks
- Bottom-Up and Top-Down Reasoning with Hierarchical Rectified Gaussians
- Predicting Localized Primordial Star Formation with Deep Convolutional Neural Networks
- Reconstruction of Simulation-Based Physical Field by Reconstruction Neural Network Method
- RDAnet: A Deep Learning Based Approach for Synthetic Aperture Radar Image Formation
- Improving compute efficacy frontiers with SliceOut