Pose Neural Fabrics Search

Abstract

Neural Architecture Search (NAS) technologies have emerged in many domains to jointly learn the architectures and weights of the neural network. However, most existing NAS works claim they are task-specific and focus only on optimizing a single architecture to replace a human-designed neural network, in fact, their search processes are almost independent of domain knowledge of the tasks. In this paper, we propose Pose Neural Fabrics Search (PoseNFS). We explore a new solution for NAS and human pose estimation task: part-specific neural architecture search, which can be seen as a variant of multi-task learning. Firstly, we design a new neural architecture search space, Cell-based Neural Fabric (CNF), to learn micro as well as macro neural architecture using a differentiable search strategy. Then, we view locating human keypoints as multiple disentangled prediction sub-tasks, and then use prior knowledge of body structure as guidance to search for multiple part-specific neural architectures for different human parts. After search, all these part-specific CNFs have distinct micro and macro architecture parameters. The results show that such knowledge-guided NAS-based architectures have obvious performance improvements to a hand-designed part-based baseline model. The experiments on MPII and MS-COCO datasets demonstrate that PoseNFS\footnote{Code is available at \url{https://github.com/yangsenius/PoseNFS}} can achieve comparable performance to some efficient and state-of-the-art methods.

arXiv comment11 pages, 8 figures, 7 tables. Code is available at https://github.com/yangsenius/PoseNFS

References in corpus (16)

• Adam: A Method for Stochastic Optimization
• Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
• MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications
• Neural Architecture Search with Reinforcement Learning
• An Overview of Multi-Task Learning in Deep Neural Networks
• Neural Architecture Search: A Survey
• Focal Loss for Dense Object Detection
• Joint Training of a Convolutional Network and a Graphical Model for Human Pose Estimation
• Associative Embedding: End-to-End Learning for Joint Detection and Grouping
• Human pose estimation via Convolutional Part Heatmap Regression
• Searching for Efficient Multi-Scale Architectures for Dense Image Prediction
• SNAS: Stochastic Neural Architecture Search
• Random Search and Reproducibility for Neural Architecture Search
• Convolutional Neural Fabrics
• Auto-DeepLab: Hierarchical Neural Architecture Search for Semantic Image Segmentation
• Feature Partitioning for Efficient Multi-Task Architectures

Cited by in corpus (6)

• Weight-Sharing Neural Architecture Search: A Battle to Shrink the Optimization Gap
• EvoPose2D: Pushing the Boundaries of 2D Human Pose Estimation using Accelerated Neuroevolution with Weight Transfer
• AutoPose: Searching Multi-Scale Branch Aggregation for Pose Estimation
• ScaleNAS: One-Shot Learning of Scale-Aware Representations for Visual Recognition
• HR-NAS: Searching Efficient High-Resolution Neural Architectures with Lightweight Transformers
• ViPNAS: Efficient Video Pose Estimation via Neural Architecture Search