1 / 73

CVPR19-Graphonomy- Universal Human Parsing via Graph Transfer Learning

Paper list:. CVPR19-Graphonomy- Universal Human Parsing via Graph Transfer Learning AAAI18-Spaital Temporal Graph Convolutional Networks for Skeleton-based Action Recognition BMVC18-Part-based Graph Convolutional Network for Action Recognition

baltimore
Download Presentation

CVPR19-Graphonomy- Universal Human Parsing via Graph Transfer Learning

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Paper list: • CVPR19-Graphonomy- Universal Human Parsing via Graph Transfer Learning • AAAI18-Spaital Temporal Graph Convolutional Networks for Skeleton-based Action Recognition • BMVC18-Part-based Graph Convolutional Network for Action Recognition • CVPR19-Actional-Structural Graph Convolutional Networks for Skeleton-based Action Recognition • CVPR19-An Attention Enhanced Graph Convolutional LSTM Network for Skeleton-Based Action Recognition • NIPS2017-Inductive Representation Learning on Large Graphs • CVPR19-Graphical Contrastive Losses for Scene Graph Generation • ECCV18-PersonLab- Person Pose Estimation and Instance Segmentation with a Bottom-Up, Part-Based, Geometric Embedding Model • MM18-RGCNN- Regularized Graph CNN for Point Cloud Segmentation • WWW19-Learning Graph Pooling and Hybrid Convolutional Operations for Text Representations • AAAI19-Multi-GCN- Graph Convolutional Networks for Multi-View Networks, with Applications to Global Poverty

  2. CVPR2019

  3. Human Parsing: huge different granularity and quantity of semantic labels • A single universal human parsing model to tackle all levels of the task(Multi-task learning) • Pretrain in one dataset, transfer to another dataset with graph transfer capability(Transfer learning)

  4. Intra-Graph Reasoning get local feature tensors from convolution layers construct graph with external structure knowledge feature maps -> graph node feature employ graph convolution three times re-project the graph nodes to image features

  5. Intra-Graph Reasoning get local feature tensors from convolution layers construct graph with external structure knowledge feature maps -> graph node feature employ graph convolution three times re-project the graph nodes to image features

  6. Inter-Graph Transfer sourcegraph ; target graph Transfer matrix: Handcraft relation: hard weight (0,1) Learnable matrix: random initialize the transfer matrix Feature similarity: computing the similarity between nodes Semantic similarity: explore linguistic knowledge

  7. AAAI18

  8. BMVC18

  9. Geometricfeatures:relativecoordinates Temporalfeatures:temporaldisplacements

  10. CVPR19

  11. Graph Convolutional Neural Network

  12. Attention Enhanced Graph Convolutional LSTM

  13. Attention Enhanced Graph Convolutional LSTM

  14. AGC-LSTMNetwork Joints Feature Representation Temporal Hierarchical Architecture: average pooling in temporal domain to increase the temporal receptive field of the top AGC-LSTM layers Learning AGC-LSTM

  15. CVPR19

  16. Spatio-Temporal GCN

  17. Actional-Structural GCN

  18. Actional-Structural GCN

  19. Actional-Structural Graph Convolution Block

  20. CVPR19

  21. NIPS2017

  22. ECCV2018

  23. Unified manner: • multi-person detection • 2D pose estimation • instance segmentation • TO DO: • identify person instance • localize facial and body keypoint • estimate instance segmentation mask

  24. Keypoint detection Produce heatmaps (one channel per keypoint), offsets(two channels per keypoint for displacements in the horizontal and vertical directions) points from the image position x to the k-th keypoint of the closest person instance j

  25. Hough voting

  26. Grouping keypoints into person detection instances • Mid-range pairwise offsets • Recurrent offset refinement • Fast greedy decoding

  27. Keypoint- and instance-level detection scoring

  28. Instance-level person segmentation points from the image position x to the position of the k-th keypoint of the corresponding instance j

More Related