Ultra-Light-Fast-Generic-Face-Detector-1MB
Ultra-Light-Fast-Generic-Face-Detector-1MB copied to clipboard
Linzaer
💎1MB lightweight face detection model (1MB轻量级人脸检测模型)
Hi, I am running on Raspberry Pi4 Bullseye 32-bit. **Input:** make -j4 **Output:** /home/pi/Ultra-Light-Fast-Generic-Face-Detector-1MB/ncnn/3rdparty/ncnn/src/layer/arm/innerproduct_arm.cpp:369:15: error: ‘activation_ss’ was not declared in this scope; did you mean ‘activation_type’? 369 | sum =...
I am running this code on Visual studio code and on it's giving me the no module named 'vision' error with python 3.7. I am not able to access the...
Thanks for this master work! About the "masked face", I have some question, hope to give me some advice please. Thanks a lot ! (1) Compare to the normal face...
thanks for your work! When I build the ncnn, there is a error like this: CMake Error at CMakeLists.txt:15 (add_subdirectory): The source directory /root/Ping/Face/Det/Ultra-Light-Fast-Generic-Face-Detector-1MB-master/ncnn/3rdparty/ncnn does not contain a CMakeLists.txt file....
我先使用onnx转成tensorflow pb，代码如下 ```python from onnx_tf.backend import prepare import onnx onnx_model = onnx.load("version-RFB-320.onnx") tf_rep = prepare(onnx_model) tf_rep.export_graph("version-RFB-320-tensorflow") ``` 然后我再将pb转为tflite ```python import tensorflow as tf saved_model_dir = 'version-RFB-320-tensorflow' # version-RFB-320-tensorflow 路径下含有pb和variables save_tf_model...
你好，我这边调用你的convert_tensorflow.py脚本，对工程中预训练模型进行转换，出现以下错误，log打出如下： Traceback (most recent call last): File "/home/wfy/project/python_project/face/ultra_face/tf/convert_tensorflow.py", line 37, in main() File "/home/wfy/project/python_project/face/ultra_face/tf/convert_tensorflow.py", line 23, in main model = create_slim_net(input_shape, base_channel, num_classes) File "/home/wfy/project/python_project/face/ultra_face/tf/model/slim_320.py", line 64, in create_slim_net center_variance,...
我使用了自己的数据集用来检测人手，并且筛选过了标签使其没有宽度或高度为0的数据，我在测试训练中将lr设置为1e-6，但它依然出现了inf的情况，请问有什么解决办法吗
Hi Linzaer. I got an error when I started testing **run_video_face_detect.py** with a mp4 video due to types of face coordinates in boxes. they were float but they need to...
