The repository contains code for training object detectors in weakly supervised setting, i.e., learning object detectors with just image level labels. This was done as part of the course 16820 (Visual Learning and Recognition).
Requirements:
- PyTorch
- Tensorflow (Used only for Tensorboard)
- Visdom
- Pillow (PIL)
The code contains simplified implementations of the following papers:
- Oquab, Maxime, et al. "Is object localization for free?-weakly-supervised learning with convolutional neural networks." Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2015.
- Bilen, Hakan, and Andrea Vedaldi. "Weakly supervised deep detection networks." Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2016.
Sample Result from 1st paper:
Sample Results from 2nd paper:
