ADfSM

(ADfSM: All-around Depth from Small Motion)

Source code and datasets for the paper:

S. Im, H. Ha, F. Rameau, H.-G. Jeon, G. Choe and I.S. Kweon, All-around Depth from Small Motion with A Spherical Panoramic Camera - ECCV 2016

• Project page

Dependency

• OpenCV
• Ceres solver

How to run

• run main.m

Important Information

Frame selection

The current implementation uses only the first 30 frames of your video clip. If you want to try with a different number of images or different sampling rate, please change 'setting.sam_rate'.

Dense matching step

We implemented the function "DenseMatching" to receive a scale for image downsampling and the number of labels for your convenience in testing. (Default: 0.5 scale and 64 labels for quick tests, you can change them ('setting.scaling, 'setting.num_label')

For the depth refinement, we utilized a tree-based depth upsampling approach [1,2].

Authors

• Sunghoon Im
• Hyowon Ha
• Francois Rameau
• Hae-Gon Jeon
• Gyeonmin Choe
• In So Kweon

© 2017 Sunghoon Im, Korea Advanced Institute of Science and Technology (KAIST)

IMPORTANT: If you use this software please cite the following in any resulting publication:

@inproceedings{im2016all,
  title={All-Around Depth from Small Motion with a Spherical Panoramic Camera},
  author={Im, Sunghoon and Ha, Hyowon and Rameau, Fran{\c{c}}ois and Jeon, Hae-Gon and Choe, Gyeongmin and Kweon, In So},
  booktitle={European Conference on Computer Vision},
  pages={156--172},
  year={2016},
  organization={Springer}
}

References

1. Yang, Qingxiong. "Stereo matching using tree filtering." IEEE transactions on pattern analysis and machine intelligence 37.4 (2015): 834-846.
2. Yang, Qingxiong. "A non-local cost aggregation method for stereo matching." Computer Vision and Pattern Recognition (CVPR), 2012 IEEE Conference on. IEEE, 2012.