ayushgaud/darknet_ros

ROS wrapper for darknet (YOLO V2) with OpenCV 3 and Kinetic support

darknet_ros's Issues

view window freezes after launching

After launching, a window appears, and weights are correctly loaded, but then the window freezes with no error message. The window appears black

What could be wrong ?

started core service [/rosout]
process[darknet_ros_wrapper-2]: started with pid [2696]
init done
layer     filters    size              input                output
    0 conv     16  3 x 3 / 1   416 x 416 x   3   ->   416 x 416 x  16
    1 max          2 x 2 / 2   416 x 416 x  16   ->   208 x 208 x  16
    2 conv     32  3 x 3 / 1   208 x 208 x  16   ->   208 x 208 x  32
    3 max          2 x 2 / 2   208 x 208 x  32   ->   104 x 104 x  32
    4 conv     64  3 x 3 / 1   104 x 104 x  32   ->   104 x 104 x  64
    5 max          2 x 2 / 2   104 x 104 x  64   ->    52 x  52 x  64
    6 conv    128  3 x 3 / 1    52 x  52 x  64   ->    52 x  52 x 128
    7 max          2 x 2 / 2    52 x  52 x 128   ->    26 x  26 x 128
    8 conv    256  3 x 3 / 1    26 x  26 x 128   ->    26 x  26 x 256
    9 max          2 x 2 / 2    26 x  26 x 256   ->    13 x  13 x 256
   10 conv    512  3 x 3 / 1    13 x  13 x 256   ->    13 x  13 x 512
   11 max          2 x 2 / 1    13 x  13 x 512   ->    13 x  13 x 512
   12 conv   1024  3 x 3 / 1    13 x  13 x 512   ->    13 x  13 x1024
   13 conv   1024  3 x 3 / 1    13 x  13 x1024   ->    13 x  13 x1024
   14 conv    125  1 x 1 / 1    13 x  13 x1024   ->    13 x  13 x 125
   15 detection
Loading weights from /home/edanweis/catkin_ws/src/darknet_ros/tiny-yolo-voc.weights...Done!

Error launching nodes

Hi @ayushgaud

When launching the nodes roslaunch darknet_ros object_detect.launch , I get:

process[darknet_ros_wrapper-2]: started with pid [26420]
/home/edanweis/catkin_ws/devel/lib/darknet_ros/darknet_ros_node: error while loading shared libraries: libopencv_highgui.so.3.2: cannot open shared object file: No such file or directory
[darknet_ros_wrapper-2] process has died [pid 26420, exit code 127, cmd /home/edanweis/catkin_ws/devel/lib/darknet_ros/darknet_ros_node /camera/image_raw:=/camera/rgb/image_raw __name:=darknet_ros_wrapper __log:=/home/edanweis/.ros/log/21ca145c-78bf-11e7-8166-74d02bc8da6b/darknet_ros_wrapper-2.log].
log file: /home/edanweis/.ros/log/21ca145c-78bf-11e7-8166-74d02bc8da6b/darknet_ros_wrapper-2*.log

Any suggestions?

ROS Kinetic, Ubuntu 16.04, CUDA 8, CUDNN 5.1

no cudnn.h - fatal error

when making in the darknet folder I get this fatal error:

In file included from ./src/gemm.c:3:0:
./src/cuda.h:15:19: fatal error: cudnn.h: No such file or directory
compilation terminated.
Makefile:116: recipe for target 'obj-cpp-shared/gemm.o' failed
make: *** [obj-cpp-shared/gemm.o] Error 1

Is there any way to fix this? thanks!

Error making darknet-cpp-shared

I get an error when attempting to make this

./src/utils.c:283:41: warning: ignoring return value of ‘char* fgets(char*, int, FILE*)’, declared with attribute warn_unused_result [-Wunused-result]
         fgets(&line[curr], readsize, fp);
                                         ^
g++  -DOPENCV `pkg-config --cflags opencv`  -DGPU -I/usr/local/cuda/include/ -DCUDNN  -Wno-write-strings -std=c++0x -Wall -Wfatal-errors  -Ofast -DOPENCV -DGPU -DCUDNN -fPIC -c ./src/cuda.c -o obj-cpp-shared/cuda.o
g++  -DOPENCV `pkg-config --cflags opencv`  -DGPU -I/usr/local/cuda/include/ -DCUDNN  -Wno-write-strings -std=c++0x -Wall -Wfatal-errors  -Ofast -DOPENCV -DGPU -DCUDNN -fPIC -c ./src/deconvolutional_layer.c -o obj-cpp-shared/deconvolutional_layer.o
g++  -DOPENCV `pkg-config --cflags opencv`  -DGPU -I/usr/local/cuda/include/ -DCUDNN  -Wno-write-strings -std=c++0x -Wall -Wfatal-errors  -Ofast -DOPENCV -DGPU -DCUDNN -fPIC -c ./src/convolutional_layer.c -o obj-cpp-shared/convolutional_layer.o
./src/convolutional_layer.c: In function ‘void cudnn_convolutional_setup(layer*)’:
./src/convolutional_layer.c:138:117: error: too few arguments to function ‘cudnnStatus_t cudnnSetConvolution2dDescriptor(cudnnConvolutionDescriptor_t, int, int, int, int, int, int, cudnnConvolutionMode_t, cudnnDataType_t)’
 nSetConvolution2dDescriptor(l->convDesc, l->pad, l->pad, l->stride, l->stride, 1, 1, CUDNN_CROSS_CORRELATION);
                                                                                                             ^
compilation terminated due to -Wfatal-errors.
Makefile:116: recipe for target 'obj-cpp-shared/convolutional_layer.o' failed
make: *** [obj-cpp-shared/convolutional_layer.o] Error 1

Running ROS Kinetic, Ubuntu 16.04, CUDA 8, CUDNN 5.1 (installed like this)

Any suggestions how to fix this?

Can modify darknet_ros subscribe the datatype of the image format ? I want to use the format of sensor_msgs/CompressedImage rather than sensor_msgs/Image