Feywell

@tangali Did you know the value of min_size? Where to find the definition of cfg['min_size]? I also want to know how to select the region proposal from selective serach method....

@tangali Maybe our question is not the same. I find my answer in code [here](https://github.com/rbgirshick/py-faster-rcnn/blob/781a917b378dbfdedb45b6a56189a31982da1b43/lib/datasets/pascal_voc.py) I just read the code too careless.

Hello. Do you find how to get the facemodel_info.mat? I want to know how to get the index of 3d landmark

Sorry. I mean，How to train a model for high resolution imgs. I want to use model for 2k resolution imgs

from Python3.5, iteritems is replaced by items.

Duplicate of #https://github.com/msracver/Deep-Feature-Flow/issues/21

I have download the model. BaiduYun:[百度云]( https://pan.baidu.com/s/1hEVPNHHSIDoKyrU_EKhA6A） Keys（密码）: 39c6

Maybe GPU memory is not enough. You can try to change KEY_FRAME_INTERVAL= below 5. I use KEY_FRAME_INTERVAL=5 is ok. Hope to useful for you!

@scharifa: Actually, it is ok when I train the code. But test phase, the error is throw out. I change the KEY_FRAME_INTERVAL in config.py and min_diff and max_diff. Hope it...

@scharifa You must encountered this error in test phase. If you run `demo.py`, you should change `demo.py` like follows: > max_data_shape = [[('data', (1, 3, max([v[0] for v in cfg.SCALES]),...