* add function wider face to coco dataset in x2coco.py

* add mtmct aic_vehicle baseline

* fix clip_bbox, add coco model filter, fix some bugs

* fix infer bugs, fix crops and mtmct vis

* fix engine tracker infer

* fix mtmct images deploy

* fix readme and cfg

* remove mtmct of cfg modeling and tracker

* fix doc cfg

* add general yolov3 for deepsort

* refine format

* fix no target infer

* fix readme and conflict

* add frame_rate for mot infer video

* add doc of frame_rate for mot video infer, test=document_fix

* add kitti metric

* clean kitti metric code

* fix kitticars doc

* remove kitti headtracking cfgs

* remove kitti headtracking cfgs, clean code

* fix datatype and ffmpeg, test=document_fix

* add soft link, test=document_fix

* add doc link, test=document_fix

* fix deploy image_list, test=document_fix

* mot video decode to images

* add infer_dir for mot

* fix export and unite pose infer

* remove task in mot source

* set image_dir

* update doc

Besides removing unnecessary imports, this also fixes the missing logger module in test_architectures.py within the static directory.

* jde fairmot export and video python infer

* add config for jde and deepsort, add engine metric tools

* add mot_metrics

* fix configs engine and others

* fix configs others

* move configs

* fix ap_per_class, fix doc

* fix test_mode, seqs

* fix test_mode to metric

* fix JDE arch metric

* clean requirement

* add to requirement

* fix sync_bn in cpu

* update doc on develop, test=document_fix

* update dependency of ppdet and paddle, test=document_fix

* update docs, test=document_fix

* fix some docs error

* Add S2ANet model for Oriented Object Detection.

* fix warning

* add Distill model
* add distill+prune

* update files

* update for ci-travis

* add solov2 dygraph model, test=dygraph

* add mbv1v3-yolov3

* add ssd mbv1v3 voc

* yolov3 mbv1v3 voc

* training log fix

* yolov3 voc fix

* yolov3 mbv1v3 training and eval, ssd fix

* add syncbn

* mbv1v3 clean code

* mbv1v3 clean codes

* remove ssd mbv1v3

* update modelzoo doc

* remove others

* fix ssd reader and prior_box

* ssd mbv1 config

* fix anchor steps and reader op

* ssdlite mbv1v3 infer and train

* update yolov3 config and modelzoo

* update yolov3_mbv3 modelzoo

* pre commit update

* update modelzoo

* fix mbv3 in yolov3

* fix precommit
Co-authored-by: Nnemonameless <nemonameless@qq.com>

* support mask_rcnn for kunlun

* minor