Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

ExtractFeatureWithCrop 异常退出错误 #188

Open
fanr520 opened this issue Feb 27, 2018 · 3 comments
Open

ExtractFeatureWithCrop 异常退出错误 #188

fanr520 opened this issue Feb 27, 2018 · 3 comments

Comments

@fanr520
Copy link

fanr520 commented Feb 27, 2018

使用特征提取的时候,出现下标越界出错,详细说明如下:
异常发生处:
文件:FaceIdentification\include\blob.h
代码行:82
代码:return data_.get()[i];
错误类型:下标i值越界
代码当前变量值:
i= 92108
count_=65536
shape_=[1,1,256,256]

上一级:
文件:\FaceIdentification\src\eltwise_net.cpp
代码行:97
dst_head[offset] = (input)[offset]
+ (bias)[bias->offset(nbn, c
bc, hbh, wbw)];

代码当前变量值:
op_="BAIS_ADDER"
scale_=-4.3160208e+008
upper_=-4.3160208e+008
lower_=-4.3160208e+008
num=1
channels=3
height=256
width=256

bn=0
bc=1
bh=1
bw=1

n=0
c=1
h=103
w=204
offset=92108

input shape_=[1,1,256,256] count_=65536
bias shape_=[1,3,256,256] count_=196608
output shape_=[1,3,256,256] count_=196608

上一级:
文件:\FaceIdentification\src\common_net.cpp
代码行:174
nets_[i]->Execute();

代码当前变量值:
nets_ size=11
i=0

上一级:
文件:\FaceIdentification\include\recognizer.h
代码行:112
net_->Execute();

代码当前变量值:
crop_blob shape_[1,1,256,256] count_=65536
src_img width=640 height=480 num_channels=1
points
[0] 289.84335 float
[1] 352.09399 float
[2] 378.71692 float
[3] 340.88025 float
[4] 340.72888 float
[5] 396.91446 float
[6] 309.39282 float
[7] 438.08136 float
[8] 373.70685 float
[9] 429.26556 float

上一级:
文件:FaceIdentification\tools\face_identification.cpp
代码行:111
recognizer->ExtractFeatureWithCrop(src_image, point_data, feats);
代码当前变量值:
src_image
width 640 int
height 480 int
num_channels 1 int
point_data
[0] 289.84335 float
[1] 352.09399 float
[2] 378.71692 float
[3] 340.88025 float
[4] 340.72888 float
[5] 396.91446 float
[6] 309.39282 float
[7] 438.08136 float
[8] 373.70685 float
[9] 429.26556 float
@seeta-wushuzhe @seetaface @jiezhang

@fanr520
Copy link
Author

fanr520 commented Feb 27, 2018

找到原因了,是因为传入的图像能道数不对。不过软件没有做相应的提示。

@ghost
Copy link

ghost commented Mar 15, 2018

我也碰到同样的问题,能不能讲下图像应该是几通道的?谢谢

@QinWinner
Copy link

通道数应该为1,要转换为通道数为1的灰度图像。

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants