Binarized Convolutional Landmark Localizers for Human Pose Estimation and Face Alignment with Limited Resources

@article{Bulat2017BinarizedCL,
  title={Binarized Convolutional Landmark Localizers for Human Pose Estimation and Face Alignment with Limited Resources},
  author={Adrian Bulat and Georgios Tzimiropoulos},
  journal={2017 IEEE International Conference on Computer Vision (ICCV)},
  year={2017},
  pages={3726-3734}
}
Our goal is to design architectures that retain the groundbreaking performance of CNNs for landmark localization and at the same time are lightweight, compact and suitable for applications with limited computational resources. [...] Key Method We exhaustively evaluate various design choices, identify performance bottlenecks, and more importantly propose multiple orthogonal ways to boost performance.Expand
Hierarchical Binary CNNs for Landmark Localization with Limited Resources
TLDR
This work is the first to study the effect of neural network binarization on localization tasks, namely human pose estimation and face alignment, and proposes a novel hierarchical, parallel and multi-scale residual architecture that yields large performance improvement over the standard bottleneck block while having the same number of parameters.
Multi-Task Head Pose Estimation in-the-Wild
TLDR
It is shown that the combination of head pose estimation and landmark-based face alignment significantly improve the performance of the former task and the proposed model outperforms the state-of-the-art in the face pose and visibility tasks.
Toward fast and accurate human pose estimation via soft-gated skip connections
TLDR
A hybrid network that combines the HourGlass and U-Net architectures which minimizes the number of identity connections within the network and increases the performance for the same parameter budget is introduced.
Designing Compact Convolutional Filters for Lightweight Human Pose Estimation
  • Shili Niu, Weihua Ou, +4 authors Wu Zeng
  • Wireless Communications and Mobile Computing
  • 2021
Existing methods for human pose estimation usually use a large intermediate tensor, leading to a high computational load, which is detrimental to resource-limited devices. To solve this problem, we
An Efficient Multitask Neural Network for Face Alignment, Head Pose Estimation and Face Tracking
TLDR
This work proposes an efficient multitask face alignment, face tracking and head pose estimation network (ATPN), which introduces a shortcut connection between shallow-layer features and deep- layer features to provide the structural information for face alignment and applies the CoordConv to the last few layers to provide coordinate information.
EfficientPose: Efficient Human Pose Estimation with Neural Architecture Search
TLDR
This paper proposes an efficient framework targeted at human pose estimation including two parts, the efficient backbone and the efficient head, by implementing the differentiable neural architecture search method and customize the backbone network design for pose estimation and reduce the computation cost with negligible accuracy degradation.
Disentangling 3D Pose in a Dendritic CNN for Unconstrained 2D Face Alignment
  • Amit Kumar, R. Chellappa
  • Computer Science
    2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition
  • 2018
TLDR
A single dendritic CNN, termed as Pose Conditioned Dendritic Convolution Neural Network (PCD-CNN), where a classification network is followed by a second and modular classification network, trained in an end to end fashion to obtain accurate landmark points is presented.
Multistage Model for Robust Face Alignment Using Deep Neural Networks
TLDR
A multistage model based on deep neural networks is proposed which takes advantage of spatial transformer networks, hourglass networks and exemplar-based shape constraints to incorporate face shape constraints so that misaligned landmarks caused by occlusions or cluttered backgrounds can be considerably improved.
Deep Evolutionary 3D Diffusion Heat Maps for Large-pose Face Alignment
TLDR
A new deep evolutionary model to integrate 3D Diffusion Heap Maps (DHM) to compensate for the invisible landmarks issue in large pose variation is proposed and a simple yet effective CNN feature is extracted and fed to recurrent neural networks for evolutionary learning.
Fine-Grained Head Pose Estimation Without Keypoints
TLDR
An elegant and robust way to determine pose is presented by training a multi-loss convolutional neural network on 300W-LP, a large synthetically expanded dataset, to predict intrinsic Euler angles directly from image intensities through joint binned pose classification and regression.
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 46 REFERENCES
Flowing ConvNets for Human Pose Estimation in Videos
TLDR
This work proposes a ConvNet architecture that is able to benefit from temporal context by combining information across the multiple frames using optical flow and outperforms a number of others, including one that uses optical flow solely at the input layers, one that regresses joint coordinates directly, and one that predicts heatmaps without spatial fusion.
Human Pose Estimation via Convolutional Part Heatmap Regression
TLDR
A CNN cascaded architecture specifically designed for learning part relationships and spatial context, and robustly inferring pose even for the case of severe part occlusions is proposed, and achieves top performance on the MPII and LSP data sets.
Convolutional aggregation of local evidence for large pose face alignment
TLDR
Convolutional Aggregation of Local Evidence (CALE) is proposed, a Convolutional Neural Network architecture particularly designed for addressing unconstrained face alignment and very large changes in shape and appearance typically encountered in animal faces.
HyperFace: A Deep Multi-Task Learning Framework for Face Detection, Landmark Localization, Pose Estimation, and Gender Recognition
TLDR
The proposed method, HyperFace, fuses the intermediate layers of a deep CNN using a separate CNN followed by a multi-task learning algorithm that operates on the fused features to exploit the synergy among the tasks which boosts up their individual performances.
Fine-grained pose prediction, normalization, and recognition
TLDR
This work unifies steps in an end-to-end trainable network supervised by keypoint locations and class labels that localizes parts by a fully convolutional network to focus the learning of feature representations for the fine-grained classification task.
Stacked Hourglass Networks for Human Pose Estimation
TLDR
This work introduces a novel convolutional network architecture for the task of human pose estimation that is described as a “stacked hourglass” network based on the successive steps of pooling and upsampling that are done to produce a final set of predictions.
Aggregated Residual Transformations for Deep Neural Networks
TLDR
On the ImageNet-1K dataset, it is empirically show that even under the restricted condition of maintaining complexity, increasing cardinality is able to improve classification accuracy and is more effective than going deeper or wider when the authors increase the capacity.
2D Human Pose Estimation: New Benchmark and State of the Art Analysis
TLDR
A novel benchmark "MPII Human Pose" is introduced that makes a significant advance in terms of diversity and difficulty, a contribution that is required for future developments in human body models.
Very Deep Convolutional Networks for Large-Scale Image Recognition
TLDR
This work investigates the effect of the convolutional network depth on its accuracy in the large-scale image recognition setting using an architecture with very small convolution filters, which shows that a significant improvement on the prior-art configurations can be achieved by pushing the depth to 16-19 weight layers.
DeeperCut: A Deeper, Stronger, and Faster Multi-person Pose Estimation Model
The goal of this paper is to advance the state-of-the-art of articulated pose estimation in scenes with multiple people. To that end we contribute on three fronts. We propose (1) improved body part
...
1
2
3
4
5
...