Hypercolumns for Object Segmentation and Fine-Grained Localization

Bharath Hariharan, Pablo Arbelaez, Ross Girshick, Jitendra Malik

CVPR 2015

doi:10.1109/CVPR.2015.7298642 /cvpr/2015/hariharan2015cvpr-hypercolumns/

Abstract

Recognition algorithms based on convolutional networks (CNNs) typically use the output of the last layer as feature representation. However, the information in this layer may be too coarse to allow precise localization. On the contrary, earlier layers may be precise in localization but will not capture semantics. To get the best of both worlds, we define the hypercolumn at a pixel as the vector of activations of all CNN units above that pixel. Using hypercolumns as pixel descriptors, we show results on three fine-grained localization tasks: simultaneous detection and segmentation[20], where we improve state-of-the-art from 49.7 mean AP^r[20] to 59.0, keypoint localization, where we get a 3.3 point boost over [19] and part labeling, where we show a 6.6 point gain over a strong baseline.

PDF CVPR Semantic Scholar

Cite

Text

Hariharan et al. "Hypercolumns for Object Segmentation and Fine-Grained Localization." Conference on Computer Vision and Pattern Recognition, 2015. doi:10.1109/CVPR.2015.7298642

Markdown

[Hariharan et al. "Hypercolumns for Object Segmentation and Fine-Grained Localization." Conference on Computer Vision and Pattern Recognition, 2015.](https://mlanthology.org/cvpr/2015/hariharan2015cvpr-hypercolumns/) doi:10.1109/CVPR.2015.7298642

BibTeX

@inproceedings{hariharan2015cvpr-hypercolumns,
  title     = {{Hypercolumns for Object Segmentation and Fine-Grained Localization}},
  author    = {Hariharan, Bharath and Arbelaez, Pablo and Girshick, Ross and Malik, Jitendra},
  booktitle = {Conference on Computer Vision and Pattern Recognition},
  year      = {2015},
  doi       = {10.1109/CVPR.2015.7298642},
  url       = {https://mlanthology.org/cvpr/2015/hariharan2015cvpr-hypercolumns/}
}