"multiclass" face detection

Here is interesting conclusion.

The created detectors outperformed the opencv default xml in terms of synthesized test samples created from training samples. This shows that the training was successfully performed. However, the detector did not work well in general test samples. This might mean that the detector was over-trained or over-fitted to the specific training samples. I still don't know good parameters or training samples to generalize detectors well.

False alarm rates of all of my generated detectors were pretty low compared with the opencv default detector. I don't know which parameters are especially different. I set false alarm rate with 0.5 and this makes sense theoretically. I don't know.

Training illumination varying faces in one detector resulted in pretty poor. The generated detector became sensitive to illumination rather than robust to illumination. This detector does not detect non-illuminated normal frontal faces. This makes sense because normal frontal faces did not exist in training sets so many. Training multi-view faces in one time resulted in the same thing.

We should train different detectors for each face pose or illumination state to construct a multi-view or illumination varied face detector as Fast Multi-view Face Detection. Viola and Jones extended their work for multi-view by training 12 separated face poses detectors. To achieve rapidness, they further constructed a pose estimator by C4.5 decision tree re-using the haar-like features, they further cascaded the pose estimator and face detector (Of course, this means that if pose estimation fails, the face detection also fails).

so the problem is then we want to train haar detector on the data that have one class and big inclass variation nothing works, we must train N detectors on subclasses of the data.

How to resolve this problem? We should claster data and divide it to N subclasses and then build N detectors, but how then to make a decision based on size of the cluster or other cluster properties?