Branch: master
Find file History
Latest commit 695e6af Dec 9, 2018
Type Name Latest commit message Commit time
Failed to load latest commit information.
model add fc_vgg16 Dec 9, 2018
tools Update Jan 20, 2018

FC image retrieval

FC means using fully-connected layer as image feature. The FC image retrieval tries to provide a baseline for layer selection for image retrieval. For more details, visit to the post.

Poject structure

➜  fc_retrieval git:(master) ✗ tree -L 2
├── feats
├── model
│   ├── pool5_neural.prototxt
│   └── pool5_vgg16.prototxt
├── src
│   ├──
│   ├──
│   └──
└── tools
    └── compute_ap.cpp

Build evaluation script

The official C++ program provided by the Oxford group, compute_ap.cpp, for computing the mean average precision (mAP) on the retrieval benchmark is provided in this repository for convenience. It is modified to add an explicit include so that it can be compiled everywhere.

g++ -O compute_ap.cpp -o compute_ap

CNN Model

model Dataset used to fine-tuning layer mAP
neural.caffemodel landmark dataset (baiduyun code: 8hjn) fc6 60.2%
VGG_ILSVRC_16_layers.caffemodel - fc6 45.9%