• Corpus ID: 202577974

# Pose Neural Fabrics Search

@article{Yang2019PoseNF,
title={Pose Neural Fabrics Search},
author={Sen Yang and Wankou Yang and Zhen Cui},
journal={ArXiv},
year={2019},
volume={abs/1909.07068}
}
• Published 16 September 2019
• Computer Science
• ArXiv
Neural Architecture Search (NAS) technologies have been successfully performed for efficient neural architectures for tasks such as image classification and semantic segmentation. However, existing works implement NAS for target tasks independently of domain knowledge and focus only on searching for an architecture to replace the human-designed network in a common pipeline. \emph{Can we exploit human prior knowledge to guide NAS?} To address it, we propose a framework, named Pose Neural Fabrics…

## Figures and Tables from this paper

EfficientPose: Efficient Human Pose Estimation with Neural Architecture Search
• Computer Science
Comput. Vis. Media
• 2021
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.
AutoPose: Searching Multi-Scale Branch Aggregation for Pose Estimation
A novel neural architecture search(NAS) framework that is capable of automatically discovering multiple parallel branches of cross-scale connections towards accurate and high-resolution 2D human pose estimation and a novel bi-level optimization method is presented.
EvoPose2D: Pushing the Boundaries of 2D Human Pose Estimation using Neuroevolution
• Computer Science
ArXiv
• 2020
This work proposes a new weight transfer scheme that relaxes function-preserving mutations, enabling us to accelerate neuroevolution in a flexible manner and produces 2D human pose network designs that are more efficient and more accurate than state-of-the-art hand-designed networks.
ScaleNAS: One-Shot Learning of Scale-Aware Representations for Visual Recognition
This work presents ScaleNAS, a one-shot learning method for exploring scale-aware representations for multi-scale supernet driven by grouped sampling and evolutionary search, and uses it to create high-resolution models for two different tasks, ScaleNet-P for human pose estimation and Scale net-S for semantic segmentation.
Automated work efficiency analysis for smart manufacturing using human pose tracking and temporal action localization
• Computer Science
J. Vis. Commun. Image Represent.
• 2020
The generative adversarial networks (GANs) are explored to achieve significantly improved estimation of human body joints and the automated work efficiency analysis is able to achieve action localization with an average IoU (intersection over union) score large than 0.9.
Weight-Sharing Neural Architecture Search: A Battle to Shrink the Optimization Gap
A literature review on the application of NAS to computer vision problems is provided and existing approaches are summarized into several categories according to their efforts in bridging the gap.
EvoPose2D: Pushing the Boundaries of 2D Human Pose Estimation Using Accelerated Neuroevolution With Weight Transfer
• Computer Science
IEEE Access
• 2021
This work explores the application of neuroevolution, a form of neural architecture search inspired by biological evolution, in the design of 2D human pose networks for the first time, and produces network designs that are more efficient and more accurate than state-of-the-art hand-designed networks.
HR-NAS: Searching Efficient High-Resolution Neural Architectures with Lightweight Transformers
• Mingyu Ding, +4 authors P. Luo
• Computer Science
2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
• 2021
This work proposes a novel NAS method, called HR-NAS, which is able to find efficient and accurate networks for different tasks, by effectively encoding multiscale contextual information while maintaining high-resolution representations.
Self-Attention Network for Human Pose Estimation
• Computer Science
• 2021
This work proposes an end-to-end self-attention network (SAN) to enable an SAN for both 2D and 3D pose estimations, and designs a compatible, effective and general joint learning framework to mix up the usage of different dimension data.
Single Person Pose Estimation: A Survey
• Feng Zhang
• Computer Science
ArXiv
• 2021
This survey examines and survey all the components of a typical human pose estimation pipeline, including data augmentation, model architecture and backbone, supervision representation, post-processing, standard datasets, evaluation metrics, and envisage the future directions.

## References

SHOWING 1-10 OF 77 REFERENCES
Auto-DeepLab: Hierarchical Neural Architecture Search for Semantic Image Segmentation
• Chenxi Liu, +4 authors Li Fei-Fei
• Computer Science
2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
• 2019
This paper presents a network level search space that includes many popular designs, and develops a formulation that allows efficient gradient-based architecture search and demonstrates the effectiveness of the proposed method on the challenging Cityscapes, PASCAL VOC 2012, and ADE20K datasets.
Learning Feature Pyramids for Human Pose Estimation
• Wei Yang
• Computer Science
2017 IEEE International Conference on Computer Vision (ICCV)
• 2017
This work designs a Pyramid Residual Module (PRMs) to enhance the invariance in scales of DCNNs and provides theoretic derivation to extend the current weight initialization scheme to multi-branch network structures.
Cascaded Pyramid Network for Multi-person Pose Estimation
• Yilun Chen, Jian Sun
• Computer Science
2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition
• 2018
A novel network structure called Cascaded Pyramid Network (CPN) is presented which targets to relieve the problem from these "hard" keypoints, with state-of-art results on the COCO keypoint benchmark, with average precision at 73.0.
Convolutional Neural Fabrics
• Computer Science
NIPS
• 2016
A "fabric" that embeds an exponentially large number of architectures that is competitive with the state of the art for image classification on MNIST and CIFAR10, and for semantic segmentation on the Part Labels dataset.
NAS-FPN: Learning Scalable Feature Pyramid Architecture for Object Detection
• Computer Science
2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
• 2019
The adopted Neural Architecture Search is adopted and a new feature pyramid architecture in a novel scalable search space covering all cross-scale connections is discovered, named NAS-FPN, which achieves better accuracy and latency tradeoff compared to state-of-the-art object detection models.
Searching for Efficient Multi-Scale Architectures for Dense Image Prediction
This work constructs a recursive search space for meta-learning techniques for dense image prediction focused on the tasks of scene parsing, person-part segmentation, and semantic image segmentation and demonstrates that even with efficient random search, this architecture can outperform human-invented architectures.
Learning Transferable Architectures for Scalable Image Recognition
• Computer Science
2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition
• 2018
This paper proposes to search for an architectural building block on a small dataset and then transfer the block to a larger dataset and introduces a new regularization technique called ScheduledDropPath that significantly improves generalization in the NASNet models.
Convolutional Pose Machines
• Computer Science
2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
• 2016
This work designs a sequential architecture composed of convolutional networks that directly operate on belief maps from previous stages, producing increasingly refined estimates for part locations, without the need for explicit graphical model-style inference in structured prediction tasks such as articulated pose estimation.
Human Pose Estimation via Convolutional Part Heatmap Regression
• Computer Science
ECCV
• 2016
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.
Hierarchical Contextual Refinement Networks for Human Pose Estimation
• Medicine, Computer Science
IEEE Transactions on Image Processing
• 2019
A new hierarchical contextual refinement network (HCRN) to robustly predict human poses in an efficient manner, where human body joints of different complexities are processed at different layers in a context hierarchy through effectively exploiting informative contexts provided in the previous layer.