xiaobin1231 / fall-detection-by-yolov3-and-liteflownet Goto Github PK

Why does the video play so slowly

codec: MJPG
input video: .mp4

why can not save result video

your work is cool!
excuse me, I ran your project。But, my results:


fall show 'normal'. So can you tell me how improve it?

Is it possible to provide fall detection dataset？

Can you update the detail on how to retrain the model.
Thank you very much!

python detect.py --source "/home/ubuntu/MyFiles/2.mp4" --weights"/home/ubuntu/MyFiles/yolov3.pt" --LiteFlowNet_weights "/home/ubuntu/MyFiles/LiteFlowNet.pt"
Namespace(LiteFlowNet_weights='/home/ubuntu/MyFiles/LiteFlowNet.pt', agnostic_nms=False, cfg='cfg/yolov3.cfg', classes=None, conf_thres=0.3, device='', fourcc='mp4v', half=False, img_size=512, iou_thres=0.5, names='data/coco.names', output='output', save_txt=False, source='/home/ubuntu/MyFiles/2.mp4', view_img=True, weights='/home/ubuntu/MyFiles/yolov3.pt')
Using CUDA device0 _CudaDeviceProperties(name='GeForce GTX 1050 Ti', total_memory=4040MB)

video 1/1 (1/29) /home/ubuntu/MyFiles/2.mp4: video 1/1 (2/29) /home/ubuntu/MyFiles/2.mp4: Traceback (most recent call last):
File "detect.py", line 232, in
detect()
File "detect.py", line 104, in detect
wholeflows = batch_estimate(flows_esti, LiteFlowNet)
File "/home/ubuntu/MyFiles/Fall-Detection-By-YOLOV3-and-LiteFlowNet-master/LiteFlowNet/run.py", line 386, in batch_estimate
tensorFlow = torch.nn.functional.interpolate(input=moduleNetwork(tensorPreprocessedFirst, tensorPreprocessedSecond),
File "/usr/local/lib/python3.7/dist-packages/torch/nn/modules/module.py", line 532, in call
result = self.forward(*input, **kwargs)
File "/home/ubuntu/MyFiles/Fall-Detection-By-YOLOV3-and-LiteFlowNet-master/LiteFlowNet/run.py", line 327, in forward
tensorFlow = self.moduleMatching[intLevel](tensorFirst[intLevel], tensorSecond[intLevel], tensorFeaturesFirst[intLevel], tensorFeaturesSecond[intLevel], tensorFlow)
File "/usr/local/lib/python3.7/dist-packages/torch/nn/modules/module.py", line 532, in call
result = self.forward(*input, **kwargs)
File "/home/ubuntu/MyFiles/Fall-Detection-By-YOLOV3-and-LiteFlowNet-master/LiteFlowNet/run.py", line 178, in forward
tensorCorrelation = torch.nn.functional.leaky_relu(input=correlation.FunctionCorrelation(tensorFirst=tensorFeaturesFirst, tensorSecond=tensorFeaturesSecond, intStride=1), negative_slope=0.1, inplace=False)
AttributeError: module 'correlation' has no attribute 'FunctionCorrelation'

hi i get this error can you help me plase

Traceback (most recent call last):
File "detect.py", line 8, in
from LiteFlowNet.run import Network, batch_estimate
File "/home/fal/Downloads/Fall-Detection-By-YOLOV3-and-LiteFlowNet-master/LiteFlowNet/run.py", line 21, in
assert(int(str('').join(torch.version.split('.')[0:3])) >= 41) # requires at least pytorch version 0.4.1
ValueError: invalid literal for int() with base 10: '170+cu110'