* add voc_eval and yolo_darknet_voc

* add yolov3_darknet_voc in MODEL_ZOO

* fix default im_size

* fix MODEL_ZOO note

* fix is_bbox_normalized

* extract map to map_utils.py

* update yolov3_dorknet_voc mixup

* add yolov3_r34_voc

* add yolov3_mobilenet_v1_voc

* fix drop empty in VAL mode

* use cfg.num_classes

* assert metric valid

* enable difficulty can be None

* add comment for bbox_eval

* num_classes in retinanet

* fix infer doc and config default

* refine dataset exists

* fix None proc

* Fix some docs
* Unify COCO and VOC
* Change PASCAL to Pascal
* Unify dataset/coco

* Rename object_detection to PaddleDetection
* Small fix for doc

* Clean up config files

- reorder: batch_size -> data_set -> sample_transforms -> batch_transforms -> worker -> other
- unify bool values, lower case true/false
- remove `null` values
- remove `shuffle` settings, covered by default values

* Update ResNet50 pretrained weight url

* Remove `use_padded_im_info` settings, covered by default value

* test=develop add retinanet_r101 and delete needless configures in data feeder

* test=develop
add retinanet_r101 and delete needless configures in data feeder

* add VOC visualize

* fixn ssd_mobilenet_v1_voc.yml

* use default label

* clean TestFeed dataset config

* fix voc default label

* fix format

* fix as review

* revert voc default

* use defult label for all

* enable batch size != 1

* refine infer with args

* remove samples

* fix as review

* refine code

* refine args

* move get_test_images to infer.py

* add visualize log

* fix images = []

* fix args

* refine infer.py

* fix yolov3_r34.yml

* Unified object detection framework based on PaddlePaddle.
* Include algo: Faster, Mask, FPN, Cascade, RetinaNet, Yolo v3, SSD.