Segmentation Fault while running demo

[l0gicforce]

2018-07-06 09:39:07.632390: I tensorflow/core/platform/cpu_feature_guard.cc:140] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2 FMA Loading model weights from data/demo_models/vgg16_fcn_color_single_frame_2d_pose_add_lov_iter_160000.ckpt data/demo_images/000004-color.png ./experiments/scripts/demo.sh: line 17: 3175 Segmentation fault (core dumped) ./tools/demo.py --gpu 0 --network vgg16_convs --model data/demo_models/vgg16_fcn_color_single_frame_2d_pose_add_lov_iter_160000.ckpt --imdb lov_keyframe --cfg experiments/cfgs/lov_color_2d.yml --rig data/LOV/camera.json --cad data/LOV/models.txt --pose data/LOV/poses.txt --background data/cache/backgrounds.pkl

Hi, while running the demo it loads the network and starts with the first image and terminates with a segmentation fault as described above.

I am running it on: 32GB RAM Nvidia GTX 1070 Ubuntu 16.04 CUDA 8.0 cudNN 7.05 Tensorflow 1.8

It was tested with multiple tensorflow versions from 1.4 to 1.8 with no success

[Assets01]

Hi! I met the same error when I run the demo.sh, test.sh and the train.sh. I have no clue to solve it. Have you figured it out ？If you do, please let me know.

[Guru-Uni-siegen]

hough_voting_gpu_op.cc has a problem. In-Line 756-757 segmentation fault occurs because dx.size() and dy.size() can be zero some time. I added a simple if statement to check the condition it works fine now.