Skip to content

haoxuhao/Image-co-localization

Repository files navigation

Image co-localization

This repo is initially based on the Deep-Descriptor-Transforming.

New features

  • Spherical cluster method added paper.
  • Test and evaluation script added.
  • VOC & Object discovery dataset preprocessing.

Visualizations

DDTplus:

CorLoc

Object discovery

impl Airplane Car Horse Mean
here(DDT) 82.9 70.8 81.7 78.5
paper 91.46 95.51 77.42 88.13

VOC 07

impl aeroplane bicycle bird boat bottle bus car cat chair cow diningtable dog horse motorbike person pottedplant sheep sofa train tvmonitor Mean
here(DDT) 61.4 52.2 64.2 16.3 23.9 61.5 66.5 87.0 18.9 54.0 22.4 74.0 76.2 68.4 39.7 19.4 30.9 47.2 68.8 23.4 48.8
paper 67.3 63.3 61.3 22.7 8.5 64.8 57.0 80.5 9.4 49.0 22.5 72.6 73.8 69.9 7.2 15.0 35.3 54.7 75.0 29.4 46.9

VOC 12

impl aeroplane bicycle bird boat bottle bus car cat chair cow diningtable dog horse motorbike person pottedplant sheep sofa train tvmonitor Mean
here(DDT) 74.8 67.4 60.3 21.6 28.3 81.6 64.0 81.8 31.9 55.8 33.3 75.4 72.4 71.0 37.7 27.0 44.1 51.4 43.2 24.2 52.4
paper 76.7 67.1 57.9 30.5 13.0 81.9 48.3 75.7 18.4 48.8 27.5 71.8 66.8 73.5 6.1 18.5 38.0 54.7 78.6 34.6 49.4

About

Image co-localization

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published