I'm sure I have installed correct version of dependencies ,including Eigen,Sophus,Pangolin and so on.I'm not meeting Errors when "sh make.sh",I also succeed in installing KinectFusion ,that part is OK.But When I download and test the trained model Error occurs.It said "Framebuffer with requested attributes not available. Using available framebuffer. You may see visual artifacts.created window" . The details are as follows:
exec
++ tee -a experiments/logs/rgbd_scene_multi_rgbd_test.txt.2019-01-11_15-03-36
echo Logging output to experiments/logs/rgbd_scene_multi_rgbd_test.txt.2019-01-11_15-03-36
Logging output to experiments/logs/rgbd_scene_multi_rgbd_test.txt.2019-01-11_15-03-36
T_dc: 1 0 0 0
0 1 0 0
0 0 1 0
0 0 0 1
Framebuffer with requested attributes not available. Using available framebuffer. You may see visual artifacts.created window
./experiments/scripts/rgbd_scene_multi_rgbd_test.sh: 行 25: 15989 段错误 (核心已转储) ./tools/test_net.py --gpu 0 --network vgg16 --model data/fcn_models/rgbd_scene/vgg16_fcn_rgbd_multi_frame_rgbd_scene_iter_40000.ckpt --imdb rgbd_scene_val --cfg experiments/cfgs/rgbd_scene_multi_rgbd.yml --rig data/RGBDScene/camera.json --kfusion 1
"段错误(核心转存储)" means “segmentation fault (core dumped)”. I check the code and I have found in fact Error is at the line 309 of test.py :
if is_kfusion:
KF = kfusion.PyKinectFusion(rig_filename)
But the KinectFusion has been installed correctly so I don't know what causes the Error.
Probable Reason may be:
1.I use RTX 2080 ,with Cuda 8.0 ,Cudnn 6.0.The device is not compatible.
2.I'm using a new computer,I have not installed OpenCV and other libraries in my computer yet.
@yuxng any suggestions? I'm vert interested in this superb project and thanks a lot .
I'm sure I have installed correct version of dependencies ,including Eigen,Sophus,Pangolin and so on.I'm not meeting Errors when "sh make.sh",I also succeed in installing KinectFusion ,that part is OK.But When I download and test the trained model Error occurs.It said "Framebuffer with requested attributes not available. Using available framebuffer. You may see visual artifacts.created window" . The details are as follows:
:~/DA-RNN-master$ ./experiments/scripts/rgbd_scene_multi_rgbd_test.sh 0
vgg16
in training 2019-01-11 15:03:37.225146: I tensorflow/core/platform/cpu_feature_guard.cc:137] Your CPU supports instructions that this TensorFlow binary was not compiled to use: SSE4.1 SSE4.2 AVX AVX2 FMA 2019-01-11 15:03:37.339921: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:892] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero 2019-01-11 15:03:37.340290: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1030] Found device 0 with properties: name: GeForce RTX 2080 major: 7 minor: 5 memoryClockRate(GHz): 1.71 pciBusID: 0000:01:00.0 totalMemory: 7.76GiB freeMemory: 7.48GiB 2019-01-11 15:03:37.340301: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1120] Creating TensorFlow device (/device:GPU:0) -> (device: 0, name: GeForce RTX 2080, pci bus id: 0000:01:00.0, compute capability: 7.5) Loading model weights from data/fcn_models/rgbd_scene/vgg16_fcn_rgbd_multi_frame_rgbd_scene_iter_40000.ckpt rgbd_scene_val aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa {"down":[0,1,0],"forward":[0,0,1],"height":480,"param_names":["fu","fv","u0","v0","k1","k2","k3"],"params":[570.29999999999995,570.29999999999995,320,240,0,0,0],"right":[1,0,0],"serialno":"34178534347","type":"Poly3","width":640} "Poly3" params: 570.29999999999995 570.29999999999995 320 240 0 0 0pose: [ [ 1, 0, 0, 0 ], [ 0, 1, 0, 0 ], [ 0, 0, 1, 0 ] ]
{"down":[0,1,0],"forward":[0,0,1],"height":480,"param_names":["fu","fv","u0","v0","k1","k2","k3"],"params":[570.29999999999995,570.29999999999995,320,240,0,0,0],"right":[1,0,0],"serialno":"34178534347","type":"Poly3","width":640} "Poly3" params: 570.29999999999995 570.29999999999995 320 240 0 0 0
pose: [ [ 1, 0, 0, 0 ], [ 0, 1, 0, 0 ], [ 0, 0, 1, 0 ] ]
T_dc: 1 0 0 0 0 1 0 0 0 0 1 0 0 0 0 1 Framebuffer with requested attributes not available. Using available framebuffer. You may see visual artifacts.created window ./experiments/scripts/rgbd_scene_multi_rgbd_test.sh: 行 25: 15989 段错误 (核心已转储) ./tools/test_net.py --gpu 0 --network vgg16 --model data/fcn_models/rgbd_scene/vgg16_fcn_rgbd_multi_frame_rgbd_scene_iter_40000.ckpt --imdb rgbd_scene_val --cfg experiments/cfgs/rgbd_scene_multi_rgbd.yml --rig data/RGBDScene/camera.json --kfusion 1
"段错误(核心转存储)" means “segmentation fault (core dumped)”. I check the code and I have found in fact Error is at the line 309 of test.py : if is_kfusion: KF = kfusion.PyKinectFusion(rig_filename) But the KinectFusion has been installed correctly so I don't know what causes the Error. Probable Reason may be: 1.I use RTX 2080 ,with Cuda 8.0 ,Cudnn 6.0.The device is not compatible. 2.I'm using a new computer,I have not installed OpenCV and other libraries in my computer yet.
@yuxng any suggestions? I'm vert interested in this superb project and thanks a lot .