We propose a deep convolutional neural network architecture codenamed Inception that achieves the new state of the art for classification and detection in the ImageNet Large-Scale Visual Recognition Challenge 2014 (ILSVRC14). The main hallmark of this architecture is the improved utilization of the computing resources inside the network. By a carefully crafted design, we increased the depth and width of the network while keeping the computational budget constant. To optimize quality, the architectural decisions were based on the Hebbian principle and the intuition of multi-scale processing. One particular incarnation used in our submission for ILSVRC14 is called GoogLeNet, a 22 layers deep network, the quality of which is assessed in the context of classification and detection.

/pdf/going-deeper-with-convolutions-1yobw2o2ds.pdf

Going deeper with convolutions

This work introduces a novel convolutional network architecture for the task of human pose estimation. Features are processed across all scales and consolidated to best capture the various spatial relationships associated with the body. We show how repeated bottom-up, top-down processing used in conjunction with intermediate supervision is critical to improving the performance of the network. We refer to the architecture as a “stacked hourglass” network based on the successive steps of pooling and upsampling that are done to produce a final set of predictions. State-of-the-art results are achieved on the FLIC and MPII benchmarks outcompeting all recent methods.

/pdf/stacked-hourglass-networks-for-human-pose-estimation-dlf71xnitp.pdf

Stacked Hourglass Networks for Human Pose Estimation

Dynamics of human body skeletons convey significant information for human action recognition. Conventional approaches for modeling skeletons usually rely on hand-crafted parts or traversal rules, thus resulting in limited expressive power and difficulties of generalization. In this work, we propose a novel model of dynamic skeletons called Spatial-Temporal Graph Convolutional Networks (ST-GCN), which moves beyond the limitations of previous methods by automatically learning both the spatial and temporal patterns from data. This formulation not only leads to greater expressive power but also stronger generalization capability. On two large datasets, Kinetics and NTU-RGBD, it achieves substantial improvements over mainstream methods.

/pdf/spatial-temporal-graph-convolutional-networks-for-skeleton-1mh8rphegh.pdf

Spatial Temporal Graph Convolutional Networks for Skeleton-Based Action Recognition

This work introduces a novel convolutional network architecture for the task of human pose estimation. Features are processed across all scales and consolidated to best capture the various spatial relationships associated with the body. We show how repeated bottom-up, top-down processing used in conjunction with intermediate supervision is critical to improving the performance of the network. We refer to the architecture as a "stacked hourglass" network based on the successive steps of pooling and upsampling that are done to produce a final set of predictions. State-of-the-art results are achieved on the FLIC and MPII benchmarks outcompeting all recent methods.

Learning based methods have shown very promising results for the task of depth estimation in single images. However, most existing approaches treat depth prediction as a supervised regression problem and as a result, require vast quantities of corresponding ground truth depth data for training. Just recording quality depth data in a range of environments is a challenging problem. In this paper, we innovate beyond existing approaches, replacing the use of explicit depth data during training with easier-to-obtain binocular stereo footage. We propose a novel training objective that enables our convolutional neural network to learn to perform single image depth estimation, despite the absence of ground truth depth data. Ex-ploiting epipolar geometry constraints, we generate disparity images by training our network with an image reconstruction loss. We show that solving for image reconstruction alone results in poor quality depth images. To overcome this problem, we propose a novel training loss that enforces consistency between the disparities produced relative to both the left and right images, leading to improved performance and robustness compared to existing approaches. Our method produces state of the art results for monocular depth estimation on the KITTI driving dataset, even outperforming supervised methods that have been trained with ground truth depth.

/pdf/unsupervised-monocular-depth-estimation-with-left-right-1s6azbk2bd.pdf

Unsupervised Monocular Depth Estimation with Left-Right Consistency

We propose a new method to quickly and accurately predict human pose---the 3D positions of body joints---from a single depth image, without depending on information from preceding frames. Our approach is strongly rooted in current object recognition strategies. By designing an intermediate representation in terms of body parts, the difficult pose estimation problem is transformed into a simpler per-pixel classification problem, for which efficient machine learning techniques exist. By using computer graphics to synthesize a very large dataset of training image pairs, one can train a classifier that estimates body part labels from test images invariant to pose, body shape, clothing, and other irrelevances. Finally, we generate confidence-scored 3D proposals of several body joints by reprojecting the classification result and finding local modes.The system runs in under 5ms on the Xbox 360. Our evaluation shows high accuracy on both synthetic and real test sets, and investigates the effect of several training parameters. We achieve state-of-the-art accuracy in our comparison with related work and demonstrate improved generalization over exact whole-skeleton nearest neighbor matching.

/pdf/real-time-human-pose-recognition-in-parts-from-single-depth-28r3ot7giz.pdf

Real-time human pose recognition in parts from single depth images

We describe two new approaches to human pose estimation. Both can quickly and accurately predict the 3D positions of body joints from a single depth image without using any temporal information. The key to both approaches is the use of a large, realistic, and highly varied synthetic set of training images. This allows us to learn models that are largely invariant to factors such as pose, body shape, field-of-view cropping, and clothing. Our first approach employs an intermediate body parts representation, designed so that an accurate per-pixel classification of the parts will localize the joints of the body. The second approach instead directly regresses the positions of body joints. By using simple depth pixel comparison features and parallelizable decision forests, both approaches can run super-real time on consumer hardware. Our evaluation investigates many aspects of our methods, and compares the approaches to each other and to the state of the art. Results on silhouettes suggest broader applicability to other imaging modalities.

https://www.microsoft.com/en-us/research/wp-content/uploads/2016/02/main-39.pdf

Richard Moore

Papers

Real-time human pose recognition in parts from single depth images

Efficient Human Pose Estimation from Single Depth Images