Pose Induction.

Shubham Tulsiani, Joao Carreira and Jitendra Malik. Pose Induction for Novel Object Categories. In ICCV, 2015.

0) Setup

• Download the code git clone --recursive https://github.com/shubhtuls/poseInduction

• We first need to download the required datasets (PASCAL VOC and PASCAL3D+) and additional annotations. In addition, we also need to reorganize some data. To do this automatically, run bash initSetup.sh

• Edit the required paths in 'startup.m', specially if you've used a local copy of some data instead of downloading via initSetup.sh

• Compile external/caffe (this is a slightly modified and outdated version of the original). Sample compilation instructions are provided below. In case of any issues, refer to the installation instructions on the caffe website.

cd external/caffe
cp Makefile.config.example Makefile.config
make -j 8
#edit MATLAB_DIR in Makefile.config
make matcaffe pycaffe
cd ../..

1) Demo

• Initialize matlab in the root directory of the code.

• Run startup; demo();. this will download our pretrained model and demonstrate predicted pose for a few images. Note that all the object classes in the demo images are novel (except for car which serves as a sanity check).

2) Training Models and Reproducing Experiments

This part of the codebase is still under construction. We'll update the instructions shortly.