https://github.com/xiaolonw/adversarial-frcnn
Object detection requires the ability to be robust to illumination, deformation, occlusion and intra-class variations.
data-driven strategy – collect large-scale datasets which have object instances under different conditions.
The hope is that these examples capture all possible variations of a visual concept and the classifier can then effectively model invariance to them.
long tail : How can we sample such occlusions and deformations which lie on the tail?