enpeizhao / cvprojects Goto Github PK

您好呀，从抖音关注过来的，但是没找到虚拟人项目的源码，可以分享下嘛？谢谢~

当多个工人距离较近时，安全帽和反光背心按 IoU 确定属于哪个工人存在BUG，导致部分安全帽和反光背心不能被检测。如下图：

在static文件夹下想export模型，报错：Traceback (most recent call last):
File "tools/export_model.py", line 53, in
raise e
File "tools/export_model.py", line 40, in
from ppdet.utils.export_utils import save_infer_model, dump_infer_config
ModuleNotFoundError: No module named 'ppdet.utils.export_utils'
不知道咋解决，请求帮助
输入命令是这个：python tools/export_model.py -c ./configs/rcnn_enhance/generic/cascade_rcnn_dcn_r101_vd_fpn_gen_server_side.yml --output_dir=./inference_model -o weights=./weights/cascade_rcnn_dcn_r101_vd_fpn_gen_server_side.pdparams
不知道能否分享一份生成的infer_cfg.yml, model.pdiparams, model.pdiparams.info, model.pdmodel吗

从B站过来, 演示视频都是发在抖音(流量很高), 能否同步发到B站, 谢谢.

这个库是已经编译成功了，但是再往后进行的时候，却提示楚错了。说为定义，但明明存在这个文件，也有这些函数的定义！哪位遇到过，麻烦给指条明路，谢谢！

/usr/bin/ld: streamer/libstreamer.so: undefined reference to avformat_network_init' /usr/bin/ld: streamer/libstreamer.so: undefined reference to avformat_network_deinit'
/usr/bin/ld: streamer/libstreamer.so: undefined reference to avcodec_close' /usr/bin/ld: streamer/libstreamer.so: undefined reference to avcodec_find_encoder'
/usr/bin/ld: streamer/libstreamer.so: undefined reference to avformat_write_header' /usr/bin/ld: streamer/libstreamer.so: undefined reference to avio_open2'
/usr/bin/ld: streamer/libstreamer.so: undefined reference to av_frame_free' /usr/bin/ld: streamer/libstreamer.so: undefined reference to av_log_set_level'
/usr/bin/ld: streamer/libstreamer.so: undefined reference to avcodec_open2' /usr/bin/ld: streamer/libstreamer.so: undefined reference to av_dict_set'
/usr/bin/ld: streamer/libstreamer.so: undefined reference to av_interleaved_write_frame' /usr/bin/ld: streamer/libstreamer.so: undefined reference to avformat_alloc_output_context2'
/usr/bin/ld: streamer/libstreamer.so: undefined reference to av_image_get_buffer_size' /usr/bin/ld: streamer/libstreamer.so: undefined reference to avcodec_send_frame'
/usr/bin/ld: streamer/libstreamer.so: undefined reference to av_register_all' /usr/bin/ld: streamer/libstreamer.so: undefined reference to sws_freeContext'
/usr/bin/ld: streamer/libstreamer.so: undefined reference to av_init_packet' /usr/bin/ld: streamer/libstreamer.so: undefined reference to av_dict_free'
/usr/bin/ld: streamer/libstreamer.so: undefined reference to avcodec_get_name' /usr/bin/ld: streamer/libstreamer.so: undefined reference to av_mallocz'
/usr/bin/ld: streamer/libstreamer.so: undefined reference to sws_scale' /usr/bin/ld: streamer/libstreamer.so: undefined reference to avcodec_parameters_from_context'
/usr/bin/ld: streamer/libstreamer.so: undefined reference to av_dump_format' /usr/bin/ld: streamer/libstreamer.so: undefined reference to avio_close'
/usr/bin/ld: streamer/libstreamer.so: undefined reference to avformat_new_stream' /usr/bin/ld: streamer/libstreamer.so: undefined reference to av_rescale_q'
/usr/bin/ld: streamer/libstreamer.so: undefined reference to avcodec_receive_packet' /usr/bin/ld: streamer/libstreamer.so: undefined reference to av_frame_alloc'
/usr/bin/ld: streamer/libstreamer.so: undefined reference to av_image_fill_arrays' /usr/bin/ld: streamer/libstreamer.so: undefined reference to avcodec_alloc_context3'
/usr/bin/ld: streamer/libstreamer.so: undefined reference to avcodec_free_context' /usr/bin/ld: streamer/libstreamer.so: undefined reference to sws_getContext'
/usr/bin/ld: streamer/libstreamer.so: undefined reference to avformat_free_context' /usr/bin/ld: streamer/libstreamer.so: undefined reference to av_packet_unref'

根据代码参考的出处https://www.geeksforgeeks.org/check-if-two-given-line-segments-intersect/，
似乎这一行应该改为return true；

你好，请问人员越界及聚众项目的校准数据集列表文件是github release中的哪个文件呀？看起来都不像呀，

另外能提供下人员越界及聚众的那段视频文件嘛，麻烦大佬了！

请问这里在生成蒙版文件时，每一帧都显示 第x帧无数据 是为什么？：
https://github.com/enpeizhao/CVprojects/blob/main/codes/15.%E9%98%B2%E6%8C%A1%E5%BC%B9%E5%B9%95/README.md#22%E7%94%9F%E6%88%90%E8%92%99%E7%89%88%E6%96%87%E4%BB%B6

(NotFound) Cannot open file ./baidu_pp_detection/models/cascade_rcnn_dcn_r101_vd_fpn_gen_server_side_model_, please confirm whether the file is normal.
[Hint: Expected static_cast(fin.is_open()) == true, but received static_cast(fin.is_open()):0 != true:1.] (at ..\paddle\fluid\inference\api\analysis_predictor.cc:1116)

Traceback (most recent call last):
File "C:\论文模型\CVprojects-main\codes\9.virtual reader\demo.py", line 515, in
control.recognize()
File "C:\论文模型\CVprojects-main\codes\9.virtual reader\demo.py", line 357, in recognize
drawInfo = DrawSomeInfo()
File "C:\论文模型\CVprojects-main\codes\9.virtual reader\demo.py", line 69, in init
self.pp_dete = Baidu_PP_Detection()
File "C:\论文模型\CVprojects-main\codes\9.virtual reader\baidu_pp_wrap.py", line 39, in init
trt_calib_mode=False)
File "baidu_pp_detection\python\infer.py", line 78, in init
trt_calib_mode=trt_calib_mode)
File "baidu_pp_detection\python\infer.py", line 359, in load_predictor
predictor = fluid.core.create_paddle_predictor(config)
RuntimeError: Unknown exception

请问这是什么原因啊

您好，我fork了代码之后本地运行报错。

google搜到的结果是 comtypes 这个包只能在windows运行，由于我对这个包的功能不太了解，想请问一下，有没有mac或者linux系统下的替代包。

I hava download "cascade_rcnn_dcn_r101_vd_fpn_gen_server_side.pdparams",but I donot find fight yml file.
I use this methods "
python tools/export_model.py -c "models\cascade_rcnn_dcn_r101_vd_fpn_gen_server_side.yml" --output_dir=./inference_model -o weights="models\cascade_rcnn_dcn_r101_vd_fpn_gen_server_side.pdparams"
" but I got this issue
"yaml.constructor.ConstructorError: could not determine a constructor for the tag '!DecodeImage'"

改动cmake时发现要依赖deepstream, 请问up可以说下如何在Windows下部署吗

请问一下swin-transformer路面分割项目的环境配置问题
Python，pytorch，mmcv-full与mmdet所使用的版本号分别是多少呢
自己调试环境总是报奇怪的错
TypeError: MaskRCNN: SwinTransformer: init() got an unexpected keyword argument 'embed_dim'
'ConfigDict' object has no attribute 'test_cfg'

您好，请问这个yolov5s_person.onnx是从哪里得到的啊？我直接从官方下了文件yolov5s.onnx似乎用不了QAQ

可以共享一下口罩规范佩戴检测中树莓派使用的系统镜像文件（.img）吗？
Opencv不太好安装，还有tflite，版本不对容易引起很多问题
希望能够共享一下

codes/28.transformer_image_caption中文看图写话 多了一个空格，在 Windows 上克隆失败。
报错
error: invalid path 'codes/28.transformer_image_caption中文看图写话 /README.md'

临时解决办法：

git config core.protectNTFS false

但是这个目录就是空的了

Win10 py3.6
没有创建方块和手指线，只开启了摄像头？

识别时出现错误

Traceback (most recent call last):
File "demo_full.py", line 479, in
faceRecognize(detector=args.detector,threshold=args.threshold, write_video = args.record,resize_w=args.w,resize_h=args.h)
File "demo_full.py", line 422, in faceRecognize
frame = updateRightInfo(frame,face_info_list,face_img_list)
File "demo_full.py", line 200, in updateRightInfo
frame[(left_y+offset_y_value):(left_y+resize_w+ offset_y_value), -(left_x+resize_w):-left_x] = face_img
ValueError: could not broadcast input array from shape (80,80,3) into shape (0,80,3)

刷新时大小不一致，请问怎么解决呢