Efficient object localization using Convolutional Networks

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 [21] 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 [20] dataset and outperforms all existing approaches on the MPII-human-pose dataset [1].

Stacked Hourglass Networks for Human Pose Estimation

“…This serves to increase efficency and reduce memory usage of their method while improving localization performance in the high precision range [16]. One consideration is that for many failure cases a refinement of position within a local window would not offer much improvement since error cases often consist of either occluded or misattributed limbs.…”

“…To improve performance at high precision thresholds the prediction is offset by a quarter of a pixel in the direction of its next highest neighbor before transforming back to the original coordinate space of the image. In MPII Human Pose, some joints do not have a corresponding [1] 76.5 59.1 Toshev et al [24] 92.3 82.0 Tompson et al [16] 93.1 89.0 Chen et al [25] 95.3 92.4 Wei et al [18] 97.6 95.0 Our model 99.0 97.0 ground truth annotation. In these cases the joint is either truncated or severely occluded, so for supervision a ground truth heatmap of all zeros is provided.…”

“…Recent pose estimation systems [15][16][17][18][19][20] have universally adopted ConvNets as their main building block, largely replacing hand-crafted features and graphical models; this strategy has yielded drastic improvements on standard benchmarks [1,21,22].…”

Multi-person Pose Estimation with Local Joint-to-Person Associations

“…While the tree-structured models provide efficient inference, they struggle to model long-range characteristics of the human body. With the progress in convolutional neural network architectures, more recent works adopt CNNs to obtain stronger part detectors but still use graphical models to obtain coherent pose estimates [4,6,7].…”

“…This is mainly due to the availability of deep learning based methods for detecting joints [1][2][3][4][5]. While earlier approaches in this direction [4,6,7] combine the body part detectors with tree structured graphical models, more recent methods [1][2][3][8][9][10] demonstrate that spatial relations between joints can be directly learned by a neural network without the need of an additional graphical model. These approaches, however, assume that only a single person is visible in the image and the location of the person is known a-priori.…”